From dohuyn2001 at hotmail.com Wed Aug 1 07:17:37 2012 From: dohuyn2001 at hotmail.com (DoHyun Kim) Date: Wed, 1 Aug 2012 05:17:37 +0000 Subject: [FieldTrip] Question about BCI2000 2D cursor task with Field trip buffer Message-ID: Hi, I'm a college student currently working on the BCI2000 2D cursor task with Field trip buffer. To moves cursors all directions (up, down, left, and right), I followed the description from the wiki that event_up = event; event_up.value = 1; event_down = event; event_down.value = -1; event_null = event; event_null.value = 0; (see http://www.bci2000.org/wiki/index.php/Programming_Tutorial:Working_with_the_FieldTrip_buffer) Currently, my cursor moves right and left with settings as event_right.value=1, event_left.value =-1. While I'm trying to move my cursor on y-axis(since I'm aiming for 2D control), I couldn't figure out how I should write the script for up & down movements. Does anyone know how to do it? Thanks. -------------- next part -------------- An HTML attachment was scrubbed... URL: From vitoria.piai at gmail.com Wed Aug 1 13:22:49 2012 From: vitoria.piai at gmail.com (=?ISO-8859-1?Q?Vit=F3ria_Magalh=E3es_Piai?=) Date: Wed, 01 Aug 2012 13:22:49 +0200 Subject: [FieldTrip] plotting mri ortho with crosshair Message-ID: <50191189.6080809@gmail.com> Hi all, I'm trying to plot an MR image (not necessarily functional data) with the ortho method and with crosshairs but the crosshairs are not there at all. I used to be able to see them before and with interactive mode I was able to go through the coordinates and still seeing where the crosshairs were. What I need in the end is to be able to see on an MRI (without functional data) with crosshairs where exactly a certain coordinate is. Any suggestions? If I remember correctly, I used to plot it using: mri = ft_read_mri(template); cfg.interactive = 'yes'; cfg.crosshair = 'yes'; ft_sourceplot(cfg, mri) Thanx a lot, Vitória -- ** Please consider the environment - do you really need to print? ** From jan.schoffelen at donders.ru.nl Wed Aug 1 13:41:41 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Wed, 1 Aug 2012 13:41:41 +0200 Subject: [FieldTrip] plotting mri ortho with crosshair In-Reply-To: <50191189.6080809@gmail.com> References: <50191189.6080809@gmail.com> Message-ID: Hi Vitoria, Problem confirmed. Could you file this as a bug on bugzilla? We can take it from there. Best, Jan-Mathijs On Aug 1, 2012, at 1:22 PM, Vitória Magalhães Piai wrote: > Hi all, > > I'm trying to plot an MR image (not necessarily functional data) with the ortho method and with crosshairs but the crosshairs are not there at all. > I used to be able to see them before and with interactive mode I was able to go through the coordinates and still seeing where the crosshairs were. > What I need in the end is to be able to see on an MRI (without functional data) with crosshairs where exactly a certain coordinate is. > Any suggestions? > > If I remember correctly, I used to plot it using: > > mri = ft_read_mri(template); > cfg.interactive = 'yes'; > cfg.crosshair = 'yes'; > ft_sourceplot(cfg, mri) > > Thanx a lot, Vitória > > -- > ** Please consider the environment - do you really need to print? ** > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Wed Aug 1 15:00:13 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Wed, 1 Aug 2012 15:00:13 +0200 Subject: [FieldTrip] plotting mri ortho with crosshair In-Reply-To: <50191189.6080809@gmail.com> References: <50191189.6080809@gmail.com> Message-ID: Hi Vitoria, After consulting with Eelke, it seems that the problem is caused by some low-level bug in the graphics handling. More to the point, there seems to be something going on with the openGL rendering in some matlab versions. If you switch to cfg.renderer = 'zbuffer' you should be able to see the cross hair (but you will loose the opacity masking for functional data). For now this is the best we can offer. Cheers, JM On Aug 1, 2012, at 1:22 PM, Vitória Magalhães Piai wrote: > Hi all, > > I'm trying to plot an MR image (not necessarily functional data) with the ortho method and with crosshairs but the crosshairs are not there at all. > I used to be able to see them before and with interactive mode I was able to go through the coordinates and still seeing where the crosshairs were. > What I need in the end is to be able to see on an MRI (without functional data) with crosshairs where exactly a certain coordinate is. > Any suggestions? > > If I remember correctly, I used to plot it using: > > mri = ft_read_mri(template); > cfg.interactive = 'yes'; > cfg.crosshair = 'yes'; > ft_sourceplot(cfg, mri) > > Thanx a lot, Vitória > > -- > ** Please consider the environment - do you really need to print? ** > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Wed Aug 1 17:55:13 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Wed, 1 Aug 2012 17:55:13 +0200 Subject: [FieldTrip] prepare neighbors template file Message-ID: Hi, I am analyzing CTF MEG data with 275 channels. I would like to know where can I find neighbors template file. Or should I create one by myself! I need this file for planar gradient calculation. Thank you in advance! All the best! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From marco.buiatti at gmail.com Wed Aug 1 18:35:53 2012 From: marco.buiatti at gmail.com (Marco Buiatti) Date: Wed, 1 Aug 2012 18:35:53 +0200 Subject: [FieldTrip] interactions between two factors In-Reply-To: References: Message-ID: Thanks Arno and Stephen for your replies. Stephen, the message you cite confirms that what I propose in my first point is correct, thanks. Any statistician willing to answer to my points 2) and 3) is welcome. Best, Marco On 27 July 2012 16:25, Stephen Politzer-Ahles wrote: > Hello Marco, > > For how to test an interaction in a 2x2 design, see > http://mailman.science.ru.nl/pipermail/fieldtrip/2011-January/003447.html > (and some additional information at > http://mailman.science.ru.nl/pipermail/fieldtrip/2010-December/003338.html, > http://mailman.science.ru.nl/pipermail/fieldtrip/2011-September/004244.html, > and Anderson & Braak 2003 in Journal of Statistical Computation & > Simulation). I'm not sure about the other two issues you mentioned, though. > > Best, > Steve > >> Message: 1 >> Date: Thu, 26 Jul 2012 12:36:39 +0200 >> From: Marco Buiatti >> To: fieldtrip at donders.ru.nl >> Subject: Re: [FieldTrip] interactions between two factors >> Message-ID: >> >> >> Content-Type: text/plain; charset=ISO-8859-1 >> >> >> Dear FieldTrippers, >> >> some time ago I have posted the message below concerning how to >> compute the statistical interaction between two factors in an EEG >> study with the FieldTrip cluster-based statistical analysis. Since I >> believe it is a problem of general interest, I was confident I would >> have received some replies, no matter how critics. But I had no reply >> and I am trying to guess why: >> >> - the problem is trivial, I should go back to my statistics books and >> solve it myself; >> - the problem is ill-posed, I should go back to my statistics books >> and reformulate it correctly; >> - the problem is tabou, no one dares commits to a solution because it >> could be a wrong one. >> - the problem is solved: I should read message number #. >> >> Thanks a lot for your feedback, >> >> Best, >> >> Marco >> >> On 25 May 2012 15:58, Marco Buiatti wrote: >> > Dear FieldTrippers, >> > >> > I am analysing an EEG study with 2x4 factors: one varies between 4 >> > parametrically varying levels (1 to 4), the second between two levels. >> > >> > I have three questions concerning the use of Fieldtrip cluster-based >> > non parametric statistical analysis in this case: >> > >> > 1) How to compute the interaction between the two factors. Let's start >> > from the simplest case of a 2x2 design, factors varying between values >> > A1 and A2 for the first factor, B1 and B2 for the second. Please tell >> > me if it is correct to compute the interaction by: >> > - computing the difference diffA=ERP(A1)-ERP(A2) separately in >> > condition B1 and B2, for every subject >> > - performing a within-subjects statistical analysis between diffA in >> > condition B1 and diffA in condition B2 (function >> > statfun_depsamplesT.m). >> > >> > 2) Now consider that factor A varies parametrically between values 1 >> > to 4. For the main effect of this factor, I have used the Fieldtrip >> > function statfun_depsamplesregrT.m and I'm satisfied with it. Is it >> > correct to compute the interaction by >> > - computing the regression >> > regrA=regression(ERP(A1),ERP(A2),ERP(A3),ERP(A4)) (computed as inside >> > function statfun_depsamplesregrT.m) separately in condition B1 and B2, >> > for every subject >> > - performing a within-subjects statistical analysis between regrA in >> > condition B1 and regrA in condition B2 (function >> > statfun_depsamplesT.m)? >> > >> > 3) Since BEFORE looking at the data (this is to prevent Eric's >> > contestation...) I expect a dipolar topography for the regression >> > (data are in average reference), I would like to combine into a joint >> > cluster negative and positive clusters. I have tried by changing >> > statfun_depsamplesregrT.m by just taking the absolute value of the >> > regression, but I get weird results (a huge, non significant cluster). >> > Is it possible that since values are now all positive, I should use a >> > different statistical test at the single bin level? Any other >> > suggestions? >> > >> > Thanks in advance for your help, >> > >> > Marco >> > >> > >> > >> > -- >> > Marco Buiatti, PhD >> > >> > CEA/DSV/I2BM / NeuroSpin >> > INSERM U992 - Cognitive Neuroimaging Unit >> > B?t 145 - Point Courrier 156 >> >> > Gif sur Yvette F-91191 FRANCE >> > Ph: +33(0)169.08.65.21 >> > Fax: +33(0)169.08.79.73 >> > E-mail: marco.buiatti at gmail.com >> > http://www.unicog.org/pm/pmwiki.php/Main/MarcoBuiatti >> > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Marco Buiatti, PhD CEA/DSV/I2BM / NeuroSpin INSERM U992 - Cognitive Neuroimaging Unit Bât 145 - Point Courrier 156 Gif sur Yvette F-91191 FRANCE Ph: +33(0)169.08.65.21 Fax: +33(0)169.08.79.73 E-mail: marco.buiatti at gmail.com http://www.unicog.org/pm/pmwiki.php/Main/MarcoBuiatti *********************************************** From polomacnenad at gmail.com Wed Aug 1 18:41:56 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Wed, 1 Aug 2012 18:41:56 +0200 Subject: [FieldTrip] MEG-anatomical MRI image coregistration Message-ID: Dear Stephen, Thank you very much for your expertise! All the best! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From lihqih at gmail.com Wed Aug 1 19:33:38 2012 From: lihqih at gmail.com (qi li) Date: Wed, 1 Aug 2012 13:33:38 -0400 Subject: [FieldTrip] mne source reconstruction Message-ID: Hi, Is there anyway to map the source space mesh back to voxel space so I can use the ft_sourceplot. After MNE source reconstruction, I have source = time: [1x399 double] pos: [8196x3 double] inside: [1x8196 double] outside: [1x0 double] method: 'average' avg: [1x1 struct] cfg: [1x1 struct] I have 8196 mesh points but how to map it back to the anatomical voxel space? Thanks! Qi From jm.horschig at donders.ru.nl Thu Aug 2 08:49:06 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Thu, 02 Aug 2012 08:49:06 +0200 Subject: [FieldTrip] prepare neighbors template file In-Reply-To: References: Message-ID: <501A22E2.6020505@donders.ru.nl> Dear Nenad, the file can be found in fieldtrip/templates/neighbours It is sufficient to specify cfg.template = 'ctf275_neighb'; or cfg.layout = 'ctf275.lay'; Best, Jörn On 8/1/2012 5:55 PM, Nenad Polomac wrote: > Hi, > I am analyzing CTF MEG data with 275 channels. I would like to know > where can I find neighbors template file. Or should I create one by > myself! I need this file for planar gradient calculation. > > Thank you in advance! > > All the best! > > Nenad > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From jm.horschig at donders.ru.nl Thu Aug 2 08:52:46 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Thu, 02 Aug 2012 08:52:46 +0200 Subject: [FieldTrip] mne source reconstruction In-Reply-To: References: Message-ID: <501A23BE.8040706@donders.ru.nl> Dear Qi, Not quite sure if I understand your request correctly, but you can just rehape: / source.avg.pow = reshape(source.avg.pow, anatomical.dim)/; Then instead of being a vector, it will become a matrix. If that's not the answer you intended to get, please phrase your question differently :) Best, Jörn On 8/1/2012 7:33 PM, qi li wrote: > Hi, > > Is there anyway to map the source space mesh back to voxel space so I > can use the ft_sourceplot. > > After MNE source reconstruction, I have > > source = > > time: [1x399 double] > pos: [8196x3 double] > inside: [1x8196 double] > outside: [1x0 double] > method: 'average' > avg: [1x1 struct] > cfg: [1x1 struct] > > I have 8196 mesh points but how to map it back to the anatomical voxel > space? Thanks! > > Qi > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 2 08:59:04 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 2 Aug 2012 08:59:04 +0200 Subject: [FieldTrip] mne source reconstruction In-Reply-To: References: Message-ID: <5A868F5B-3DD0-4383-BF64-8D53B537A128@donders.ru.nl> Hi Qi, You might want to check the function 'ft_sourceinterpolate' for this. Best, Jan-Mathijs On Aug 1, 2012, at 7:33 PM, qi li wrote: > Hi, > > Is there anyway to map the source space mesh back to voxel space so I > can use the ft_sourceplot. > > After MNE source reconstruction, I have > > source = > > time: [1x399 double] > pos: [8196x3 double] > inside: [1x8196 double] > outside: [1x0 double] > method: 'average' > avg: [1x1 struct] > cfg: [1x1 struct] > > I have 8196 mesh points but how to map it back to the anatomical voxel > space? Thanks! > > Qi > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 2 09:07:22 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 2 Aug 2012 09:07:22 +0200 Subject: [FieldTrip] mne source reconstruction In-Reply-To: <501A23BE.8040706@donders.ru.nl> References: <501A23BE.8040706@donders.ru.nl> Message-ID: <4570E04B-810A-4002-B286-561FDE5C0221@donders.ru.nl> Hi Qi, To follow up on Jörn: If you reconstructed the source activations on a mesh that represents the cortical surface, you first need to interpolate the functional data into a 3D volume, if you want to use ft_sourceplot for visualization. For this you need ft_sourceinterpolate. As an alternative, to visualize the results, you can use ft_sourcemovie. However, this function is under development so it may not give the most visually appealing results. Best, JM On Aug 2, 2012, at 8:52 AM, Jörn M. Horschig wrote: > Dear Qi, > > Not quite sure if I understand your request correctly, but you can just rehape: > source.avg.pow = reshape(source.avg.pow, anatomical.dim); > Then instead of being a vector, it will become a matrix. > > If that's not the answer you intended to get, please phrase your question differently :) > > Best, > Jörn > > On 8/1/2012 7:33 PM, qi li wrote: >> Hi, >> >> Is there anyway to map the source space mesh back to voxel space so I >> can use the ft_sourceplot. >> >> After MNE source reconstruction, I have >> >> source = >> >> time: [1x399 double] >> pos: [8196x3 double] >> inside: [1x8196 double] >> outside: [1x0 double] >> method: 'average' >> avg: [1x1 struct] >> cfg: [1x1 struct] >> >> I have 8196 mesh points but how to map it back to the anatomical voxel >> space? Thanks! >> >> Qi >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > -- > Jörn M. Horschig > PhD Student > Donders Institute for Brain, Cognition and Behaviour > Centre for Cognitive Neuroimaging > Radboud University Nijmegen > Neuronal Oscillations Group > > P.O. Box 9101 > NL-6500 HB Nijmegen > The Netherlands > > Contact: > E-Mail: jm.horschig at donders.ru.nl > Tel: +31-(0)24-36-68493 > Web: http://www.ru.nl/donders > > Visiting address: > Trigon, room 2.30 > Kapittelweg 29 > NL-6525 EN Nijmegen > The Netherlands > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 2 09:16:26 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 2 Aug 2012 09:16:26 +0200 Subject: [FieldTrip] granger causality In-Reply-To: <20120730162938.56630@gmx.net> References: <20120730162938.56630@gmx.net> Message-ID: <17A8F077-820B-48A6-A388-2B2E0B7AB002@donders.ru.nl> Hi Carina, > 1.) I know that I can calculate the appropriate model order and window length with the bsmart toolbox. But from the documentation I somehow do not get the format of data I should use for this calculation. > When I enter the time-frequency data (which I want to use to estimate granger causality) (size: 2 x 2 x freq x time points), it doesn't work. The 'it doesn't work' statement is typically not enough for us to diagnose where the problem lies (at least at which step it goes wrong). At the moment I can only assume that you use an outdated version of FieldTrip, because when I try to do TF-resolved Granger using AR-models it works. It is good that you pasted the script below, but an error message may be informative too. > Any advise? I have a sampling rate of 1000. What sort of time window and maximal model order would make sense do you think? I think that the time window is rather short: I'd rather go for 0.4 or 0.5 to begin with, but this is an empirical question. Also, shifting the windows one time step at a time does not really make sense to me, 0.01 or even 0.05 is much more memory friendly. For the model order: higher one. Some people seem to get nice results when taking a relatively high model order. Of course, for a high model order you also need longer time-windows. > 2)I would like to keep the time domain analyzing my data. What do you think would be an appropriate sliding time window using the 'ft_mvaranalysis' function? And what does it mean exactly? By choosing the model order, I am already determining the maximal time lag between the two functions. Are the values then estimated for the whole time window? No. The time window determines the length of the data segment which is used to fit an AR-model (of order X) of the data at time point Y (as defined in your toi) > 3)What do you think about statistics? Would it makes sense to use the non-parametric cluster approach to shuffle within patients and do a group analysis that way? This depends on your experimental question and you null-hypothesis. Best, Jan-Mathijs > > Thank you so much in advance! > Best, > Carina > > > As a summary,I am doing following steps with field trip: > > > cfg = []; > cfg.dftfilter ='yes' > prep_cond1{subj} = ft_preprocessing(cfg, data); > > > cfg = []; > cfg.order = 5; > cfg.toolbox = 'bsmart'; > cfg.t_ftimwin = 0.05 > cfg.toi = -1:0.001:3.5; > mdata_cond1{subj}= ft_mvaranalysis(cfg, prep_cond1{subj}); > > > cfg = []; > cfg.method = 'mvar'; > cfg.foi = 4:100; > cond1_freq{subj}= ft_freqanalysis_mvar(cfg,mdata_cond1{subj}); > > > cfg = []; > cfg.method = 'granger'; > cond1_granger{subj} = ft_connectivityanalysis(cfg, cond1_freq{subj}); > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 2 09:27:38 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 2 Aug 2012 09:27:38 +0200 Subject: [FieldTrip] Question regarding ft_connectivitysimulation In-Reply-To: <5A1787011651BC42A4D41856DBC2E0603E048720@mbox-f-3.du.deakin.edu.au> References: <5A1787011651BC42A4D41856DBC2E0603E048720@mbox-f-3.du.deakin.edu.au> Message-ID: <6BC78C56-C12D-485A-B916-170A6D0EFCD7@donders.ru.nl> Hi Imali, I am not sure whether I understand completely what you want. I guess you want to simulate time series of 'sources', that have different (more or less well defined) spectral characteristics. As of yet, ft_connectivitysimulation does not support this. You could however call ft_connectivitysimulation twice, (using cfg.method = 'inear_mix') and using different cfg.bpfreq each time. Subsequently, you could sum the resulting time courses: for k = 1:numel(data1.trial) data1.trial{k} = data1.trial{k} + data2.trial{k}; end where data1 and data2 are the output to ft_connectivitysimulation. Best, Jan-Mathijs On Jul 19, 2012, at 6:44 AM, IMALI THANUJA HETTIARACHCHI wrote: > Dear fieldTrippers, > > I am trying to generate some simulated signals from a known connectivity structure with ft_connectivitysimulation, and secondly to assign these signals to known dipole locations when using in ft_dipolesimulation. > > Is there any way I can assign a frequency to the signals in ft_connectivityanalysis, so that I can assign different known frequency signals for dipoles? > > Any help is really appreciated. > > Thank you very much. > > Kind regards > Imali > > Imali Thanuja Hettiarachchi > PhD Candidate > Centre for Intelligent Systems research > Deakin University, Geelong 3217, Australia. > Email: ith at deakin.edu.au > www.deakin.edu.au/cisr > > > > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 2 09:31:03 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 2 Aug 2012 09:31:03 +0200 Subject: [FieldTrip] About co-register the individual sudbject MRI and sensor locations In-Reply-To: <5D2CC5653262C14188174E20065A4C4EC8D457@hscmbx6.uthsc.tennessee.edu> References: <5D2CC5653262C14188174E20065A4C4EC8D457@hscmbx6.uthsc.tennessee.edu> Message-ID: <3CC4394E-B87C-46A8-80C8-B002D2FF79C5@donders.ru.nl> Hi Xiaoxiao, It looks as if your mri (from which you obtained the vol-variable) and the data in the hs_file are not expressed in the same coordinate system. The geometrical objects that you need to combine in order to compute leadfields should be defined in the same coordinate system. My guess is that you will first need to apply ft_volumerealign to your mri_nom variable, to get it back into the Bti-based coordinate system. Best, Jan-Mathijs On Jul 21, 2012, at 1:28 AM, Bai, XiaoXiao wrote: > Dear Sir/Madam, > > I am a new fieldtrip user and want to apply the beamformer method in Fieldtrip for the MEG data from 4D/BTI system. Now I can get a individual subject MRI in analysis/spm format from CURRY software. > > This is my code for create volume head model for the beamformer, > > mri_nom = ft_read_mri(mrifile); > > cfg = []; > [segmentedmri] = ft_volumesegment(cfg, mri_nom); > > cfg = []; > cfg.method = 'singleshell'; > vol = ft_prepare_headmodel(cfg, segmentedmri); > > The vol and sensor locations were displayed and attached with here. > > How can I co-register the vol with sensors based on headshpe file (hs_file) for computing lead field for next step? > > Thanks a lot. > > Best regards, > > Xiaoxiao > Division of Clinical Neurosciences, Department of Pediatrics > University of Tennessee Health Science Center, College of Medicine & > Neuroscience Institute, LeBonheur Children's Hospital > 777 Washington Avenue, P335 > Memphis, TN 38105, USA > Phone: 901-287-4612 > Fax: 901-287-5325 > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Thu Aug 2 14:41:32 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Thu, 2 Aug 2012 14:41:32 +0200 Subject: [FieldTrip] prepare neighbors template file Message-ID: Dear Jörn, Thank you very much for fast answer. I have one more problem concerning this thing. When I do ft_timelockgrandaverage I got this warning: "discarding gradiometer information because it cannot be averaged". So, I tried different ways to calculate ft_megrealign. And I am not sure at which point I should do this function. Before trial averaging in single subject or after? Kind regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From lihqih at gmail.com Thu Aug 2 17:00:50 2012 From: lihqih at gmail.com (qi li) Date: Thu, 2 Aug 2012 11:00:50 -0400 Subject: [FieldTrip] mne source reconstruction In-Reply-To: <4570E04B-810A-4002-B286-561FDE5C0221@donders.ru.nl> References: <501A23BE.8040706@donders.ru.nl> <4570E04B-810A-4002-B286-561FDE5C0221@donders.ru.nl> Message-ID: Jan and Jörn: Thanks a lot for your answers! You did understand my questions pretty well. I actually tried to use ft_sourceinterpolate to find the voxel index without success. The error message is like 'Reference to non-existent field 'sphereradius'. Error in ft_sourceinterpolate (line 203) interpmat = interp_ungridded(functional.pos, warp_apply(anatomical.transform, [X(:) Y(:) Z(:)]), ...' On Thu, Aug 2, 2012 at 3:07 AM, jan-mathijs schoffelen wrote: > Hi Qi, > > To follow up on Jörn: > > If you reconstructed the source activations on a mesh that represents the > cortical surface, you first need to interpolate the functional data into a > 3D volume, if you want to use ft_sourceplot for visualization. For this you > need ft_sourceinterpolate. > As an alternative, to visualize the results, you can use ft_sourcemovie. > However, this function is under development so it may not give the most > visually appealing results. > > Best, > > JM > > > On Aug 2, 2012, at 8:52 AM, Jörn M. Horschig wrote: > > Dear Qi, > > Not quite sure if I understand your request correctly, but you can just > rehape: > source.avg.pow = reshape(source.avg.pow, anatomical.dim); > Then instead of being a vector, it will become a matrix. > > If that's not the answer you intended to get, please phrase your question > differently :) > > Best, > Jörn > > On 8/1/2012 7:33 PM, qi li wrote: > > Hi, > > Is there anyway to map the source space mesh back to voxel space so I > can use the ft_sourceplot. > > After MNE source reconstruction, I have > > source = > > time: [1x399 double] > pos: [8196x3 double] > inside: [1x8196 double] > outside: [1x0 double] > method: 'average' > avg: [1x1 struct] > cfg: [1x1 struct] > > I have 8196 mesh points but how to map it back to the anatomical voxel > space? Thanks! > > Qi > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > -- > Jörn M. Horschig > PhD Student > Donders Institute for Brain, Cognition and Behaviour > Centre for Cognitive Neuroimaging > Radboud University Nijmegen > Neuronal Oscillations Group > > P.O. Box 9101 > NL-6500 HB Nijmegen > The Netherlands > > Contact: > E-Mail: jm.horschig at donders.ru.nl > Tel: +31-(0)24-36-68493 > Web: http://www.ru.nl/donders > > Visiting address: > Trigon, room 2.30 > Kapittelweg 29 > NL-6525 EN Nijmegen > The Netherlands > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > Jan-Mathijs Schoffelen, MD PhD > > Donders Institute for Brain, Cognition and Behaviour, > Centre for Cognitive Neuroimaging, > Radboud University Nijmegen, The Netherlands > > Max Planck Institute for Psycholinguistics, > Nijmegen, The Netherlands > > J.Schoffelen at donders.ru.nl > Telephone: +31-24-3614793 > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From nenga at gmx.net Thu Aug 2 18:51:28 2012 From: nenga at gmx.net (Carina Oehrn) Date: Thu, 02 Aug 2012 18:51:28 +0200 Subject: [FieldTrip] granger causality: model order In-Reply-To: <17A8F077-820B-48A6-A388-2B2E0B7AB002@donders.ru.nl> References: <20120730162938.56630@gmx.net> <17A8F077-820B-48A6-A388-2B2E0B7AB002@donders.ru.nl> Message-ID: <20120802165128.149370@gmx.net> Hey Jan-Mathijs, thank you very much for your response and your advise on the length of time windows! TF-resolved Granger works fine with fieldtrip now. I still have my questions regarding the choice of model order and yes, sorry, my descriptions were not detailed enough. 1) what I meant was that I have problems using the aic_test function of the bsmart toolbox to calculate the most appropriate model order. When I am trying to use maximal model orders above 15 (I have 3000 data points and I tried various different time windows), I always get the error message: ??? Error using ==> aic_test at 47 Cannot compute AIC with 'opssaic'! But I can not find the reason. Others using this function seemed to have the same problem. However, as my data was sampled at a rate of 1000 Hz and as I do not downsample in my analysis, I would like to use orders above 15 (which to my knowledge most people use- who however mostly have smaller sampling rates) to include a time period longer than 15 ms. 2) What range of model orders were you talking about when stating: > For the model order: higher one. Some people seem to get nice results when taking a relatively high model order. Or does anybody else have some experience with higher model orders? How far up do you go to include a sufficient time interval but still avoid overparameterization? Thank you very much for your help! All the best, Carina -------- Original-Nachricht -------- > Datum: Thu, 2 Aug 2012 09:16:26 +0200 > Von: jan-mathijs schoffelen > An: FieldTrip discussion list > Betreff: Re: [FieldTrip] granger causality > Hi Carina, > > > 1.) I know that I can calculate the appropriate model order and window > length with the bsmart toolbox. But from the documentation I somehow do not > get the format of data I should use for this calculation. > > When I enter the time-frequency data (which I want to use to estimate > granger causality) (size: 2 x 2 x freq x time points), it doesn't work. > > The 'it doesn't work' statement is typically not enough for us to diagnose > where the problem lies (at least at which step it goes wrong). At the > moment I can only assume that you use an outdated version of FieldTrip, because > when I try to do TF-resolved Granger using AR-models it works. It is good > that you pasted the script below, but an error message may be informative > too. > > > Any advise? I have a sampling rate of 1000. What sort of time window and > maximal model order would make sense do you think? > > I think that the time window is rather short: I'd rather go for 0.4 or 0.5 > to begin with, but this is an empirical question. Also, shifting the > windows one time step at a time does not really make sense to me, 0.01 or even > 0.05 is much more memory friendly. For the model order: higher one. Some > people seem to get nice results when taking a relatively high model order. Of > course, for a high model order you also need longer time-windows. > > > 2)I would like to keep the time domain analyzing my data. What do you > think would be an appropriate sliding time window using the 'ft_mvaranalysis' > function? And what does it mean exactly? By choosing the model order, I am > already determining the maximal time lag between the two functions. Are > the values then estimated for the whole time window? > > No. The time window determines the length of the data segment which is > used to fit an AR-model (of order X) of the data at time point Y (as defined > in your toi) > > > 3)What do you think about statistics? Would it makes sense to use the > non-parametric cluster approach to shuffle within patients and do a group > analysis that way? > > This depends on your experimental question and you null-hypothesis. > > Best, > Jan-Mathijs > > > > > Thank you so much in advance! > > Best, > > Carina > > > > > > As a summary,I am doing following steps with field trip: > > > > > > cfg = []; > > cfg.dftfilter ='yes' > > prep_cond1{subj} = ft_preprocessing(cfg, data); > > > > > > cfg = []; > > cfg.order = 5; > > cfg.toolbox = 'bsmart'; > > cfg.t_ftimwin = 0.05 > > cfg.toi = -1:0.001:3.5; > > mdata_cond1{subj}= ft_mvaranalysis(cfg, prep_cond1{subj}); > > > > > > cfg = []; > > cfg.method = 'mvar'; > > cfg.foi = 4:100; > > cond1_freq{subj}= ft_freqanalysis_mvar(cfg,mdata_cond1{subj}); > > > > > > cfg = []; > > cfg.method = 'granger'; > > cond1_granger{subj} = ft_connectivityanalysis(cfg, cond1_freq{subj}); > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > Jan-Mathijs Schoffelen, MD PhD > > Donders Institute for Brain, Cognition and Behaviour, > Centre for Cognitive Neuroimaging, > Radboud University Nijmegen, The Netherlands > > Max Planck Institute for Psycholinguistics, > Nijmegen, The Netherlands > > J.Schoffelen at donders.ru.nl > Telephone: +31-24-3614793 > From fredericroux at hotmail.de Fri Aug 3 11:38:31 2012 From: fredericroux at hotmail.de (Frederic Roux) Date: Fri, 3 Aug 2012 11:38:31 +0200 Subject: [FieldTrip] PhD position Message-ID: PhD position Beginning in October 2012, the Brain Imaging Center, Department of Neurology, Goethe-University, offers a PhD position (3years, 65% of standard weekly hours) in the project: "Spectro-temporal dynamics of sensorimotor interactions underlying lateralization of speech production" funded by the German Research Council. The successful applicant will study speech production using magnetoencephalography, electro- corticography, and computational modelling. Our interdisciplinary Emmy Noether group focuses on the question how network dynamics contribute to the generation of complex behavior and offers - besides expertise in functional neuroimaging and Cognitive Neuroscience - a creative and cooperative work environment. We are seeking for an enthusiastic thesis student with interest in the Neurosciences who has advanced skills in Matlab programming and experience in signal processing and/or brain data analysis. He/she should have a background in Natural or Applied Science or Psychology. Electronic applications (no original documents) are requested until 2012/08/31 and should be addressed to: Dr. Christian Kell, Cognitive Neuroscience Group, Brain Imaging Center and Department of Neurology, Goethe University, Schleusenweg 2-16, 60528 Frankfurt, Germany c.kell at em.uni-frankfurt.de www.brainclocks.com Pay group E13 TV-GU. Teaching is not mandatory. Goethe University is an equal opportunity employer. Applications from female candidates are encouraged. Where qualifications are equal, preference will be given to people with disabilities. -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Fri Aug 3 15:36:53 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Fri, 3 Aug 2012 15:36:53 +0200 Subject: [FieldTrip] planar gradient literature Message-ID: Hi, I would like to ask for some literature concerning planar gradient computation. I am new in MEG field and I need some more information on theoretical background of such a calculation. The stuff available in "Event related averaging and planar gradient" tutorial is useful but not very detailed. So, could you please direct me to some reference. Thank you in advance! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From Lilla.Magyari at mpi.nl Fri Aug 3 16:10:31 2012 From: Lilla.Magyari at mpi.nl (Lilla.Magyari at mpi.nl) Date: Fri, 3 Aug 2012 16:10:31 +0200 (CEST) Subject: [FieldTrip] planar gradient literature In-Reply-To: References: Message-ID: <50094.131.174.45.70.1344003031.squirrel@131.174.45.70> Hi Nenad, I usually refer to this paper when I use the planar gradient computation: Bastiaansen, M. C. M., & Knösche, T. R. (2000). MEG tangential derivative mapping applied to Event-Related Desynchronization (ERD) research. Clinical Neurophysiology, 111, 1300-1305. Best, Lilla > Hi, > > I would like to ask for some literature concerning planar gradient > computation. I am new in MEG field and I need some more information on > theoretical background of such a calculation. The stuff available in > "Event > related averaging and planar gradient" tutorial is useful but not > very detailed. So, could you please direct me to some reference. > > Thank you in advance! > > Nenad > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From polomacnenad at gmail.com Fri Aug 3 18:45:48 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Fri, 3 Aug 2012 18:45:48 +0200 Subject: [FieldTrip] planar gradient literature Message-ID: Thank you Lilla! :) All the best! -------------- next part -------------- An HTML attachment was scrubbed... URL: From juan.garciaprieto at ctb.upm.es Sun Aug 5 20:41:35 2012 From: juan.garciaprieto at ctb.upm.es (Juan Garcia-Prieto) Date: Sun, 5 Aug 2012 20:41:35 +0200 Subject: [FieldTrip] anonymize registers Message-ID: Hi everybody, First time mailing this list. great pleasure. This is Juan, from madrid's MEG lab. I've coded a little matlab-FT script which anonymizes fif files (elekta neuromag). Brief story is that when sending a .fif file somewhere else for post-processing, it might be wise to eliminate any sensible information in terms subjects name, id, etc. because of data privacy concerns. When dealing with this issue, Elekta very gently supplied a script wrapper based on regular expressions, to be run under linux. So my effort will only be another way of solving the problem. The script uses "fiff_read_tag_info" "fiff_read_tag" "fiff_make_dir_tree" and "fiff_dir_tree_find" functions (part of FT) by Matti Hamalainen. My question, is: Will this function be useful to some others? Specially taking into account I have ONLY worked with neuromag .fif files. thank you for your work. I've learned a LOT, only reading FT scripts. Best regards, Juan, -------------- next part -------------- An HTML attachment was scrubbed... URL: From kashyapmanik at gmail.com Tue Aug 7 08:38:07 2012 From: kashyapmanik at gmail.com (Manik Gupta) Date: Tue, 7 Aug 2012 07:38:07 +0100 Subject: [FieldTrip] Help needed for reading a time series data Message-ID: Dear All I am new to fieldtrip and want to use to preprocess some air pollution data using fieldtrip. The data is sampled at a frequency of 1 sample/second and currently is in a text file format. Can someone please help me how to process the text file into fieldtrip raw data structure. Thanks a lot, Manik. -------------- next part -------------- An HTML attachment was scrubbed... URL: From stan.vanpelt at fcdonders.ru.nl Tue Aug 7 09:43:38 2012 From: stan.vanpelt at fcdonders.ru.nl (Stan van Pelt) Date: Tue, 7 Aug 2012 09:43:38 +0200 (CEST) Subject: [FieldTrip] Help needed for reading a time series data In-Reply-To: Message-ID: <29876546.58683.1344325418384.JavaMail.root@sculptor.zimbra.ru.nl> Dear Manik, Great that you're trying to apply FieldTrip to non-neuroscientific data. As you probably know, FT is developed for the analysis of electrophysiological data, but that does not mean it cannot be useful to you. W.r.t. your data format, i think it is easiest if you read in the data into Matlab using a function like readtxt.m, and than put these data into a FT-type structure, and run your actual analyses. At the minimum, this structure should look like this: data = label: {187x1 cell} % channel labels (e.g. 'air pollution site 1','air pollution site 2', etc) trial: {1x266 cell} % data (Nchans*Nsamples) for each trial (if you have only one long data set, it means you have only 1 trial; if you have multiple 'recording sessions' from the same site, you would have more) time: {1x266 cell} % time axis for each trial fsample: 300 % sampling frequency It would be great if you would be willing to share your results when they are published and add a FieldTrip reference (http://www.hindawi.com/journals/cin/2011/156869/). Good luck! Stan -- Stan van Pelt, PhD Ernst Strüngmann Institute (ESI) for Neuroscience in Cooperation with Max Planck Society Deutschordenstr. 46 60528 Frankfurt, Germany Website: www.esi-frankfurt.de E-mail: stan.vanpelt at esi-frankfurt.de Tel: +49 (0)69 96769 519 Fax: +49 (0)69 96769 555 ----- Oorspronkelijk bericht ----- > Van: "Manik Gupta" > Aan: fieldtrip at science.ru.nl > Verzonden: Dinsdag 7 augustus 2012 08:38:07 > Onderwerp: [FieldTrip] Help needed for reading a time series data > Dear All > I am new to fieldtrip and want to use to preprocess some air pollution > data using fieldtrip. > The data is sampled at a frequency of 1 sample/second and currently is > in a text file format. > Can someone please help me how to process the text file into fieldtrip > raw data structure. > Thanks a lot, > Manik. > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Stan van Pelt Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen Kapittelweg 29, 6525 EN Nijmegen, Netherlands E-mail: stan.vanpelt at donders.ru.nl Website: www.ru.nl/donders/ Phone: (+31) (0)24 36 10981 Fax: (+31) (0)24 36 10989 -------------- next part -------------- An HTML attachment was scrubbed... URL: From kashyapmanik at gmail.com Tue Aug 7 09:51:31 2012 From: kashyapmanik at gmail.com (Manik Gupta) Date: Tue, 7 Aug 2012 08:51:31 +0100 Subject: [FieldTrip] Help needed for reading a time series data In-Reply-To: <29876546.58683.1344325418384.JavaMail.root@sculptor.zimbra.ru.nl> References: <29876546.58683.1344325418384.JavaMail.root@sculptor.zimbra.ru.nl> Message-ID: Dear Stan, Thanks a lot for the reply. I actually want to do a transfer entropy analysis using Trentool and for that my data needs to be in the fieldtrip raw format. I figured out how to use the data now and will get back to you in case of more queries. Just a quick check, would you be able to provide any assistance on Trentool as well. and I would surely publish the results if they are encouraging enough!! :-) Thanks once again, Manik On Tue, Aug 7, 2012 at 8:43 AM, Stan van Pelt wrote: > Dear Manik, > > Great that you're trying to apply FieldTrip to non-neuroscientific data. > As you probably know, FT is developed for the analysis of > electrophysiological data, but that does not mean it cannot be useful to > you. > W.r.t. your data format, i think it is easiest if you read in the data > into Matlab using a function like readtxt.m, and than put these data into a > FT-type structure, and run your actual analyses. At the minimum, this > structure should look like this: > > data = > > label: {187x1 cell} % channel labels (e.g. 'air pollution site 1','air pollution site 2', etc) > trial: {1x266 cell} % data (Nchans*Nsamples) for each trial (if you have only one long data set, it means you have only 1 trial; if you have multiple 'recording sessions' from the same site, you would have more) > time: {1x266 cell} % time axis for each trial > fsample: 300 % sampling frequency > > It would be great if you would be willing to share your results when they > are published and add a FieldTrip reference ( > http://www.hindawi.com/journals/cin/2011/156869/). > > Good luck! > > Stan > > -- > Stan van Pelt, PhD > > Ernst Strüngmann Institute (ESI) > for Neuroscience in Cooperation with Max Planck Society > Deutschordenstr. 46 > 60528 Frankfurt, Germany > Website: www.esi-frankfurt.de > E-mail: stan.vanpelt at esi-frankfurt.de > Tel: +49 (0)69 96769 519 > Fax: +49 (0)69 96769 555 > > > ------------------------------ > > *Van: *"Manik Gupta" > *Aan: *fieldtrip at science.ru.nl > *Verzonden: *Dinsdag 7 augustus 2012 08:38:07 > *Onderwerp: *[FieldTrip] Help needed for reading a time series data > > > Dear All > > I am new to fieldtrip and want to use to preprocess some air pollution > data using fieldtrip. > The data is sampled at a frequency of 1 sample/second and currently is in > a text file format. > Can someone please help me how to process the text file into fieldtrip raw > data structure. > > Thanks a lot, > Manik. > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > -- > Stan van Pelt > > Donders Institute for Brain, Cognition and Behaviour, Radboud University > Nijmegen > Kapittelweg 29, 6525 EN Nijmegen, Netherlands > E-mail: stan.vanpelt at donders.ru.nl > Website: www.ru.nl/donders/ > Phone: (+31) (0)24 36 10981 > Fax: (+31) (0)24 36 10989 > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -------------- next part -------------- An HTML attachment was scrubbed... URL: From johanna.zumer at donders.ru.nl Tue Aug 7 14:12:20 2012 From: johanna.zumer at donders.ru.nl (Johanna Zumer) Date: Tue, 7 Aug 2012 14:12:20 +0200 Subject: [FieldTrip] mne source reconstruction In-Reply-To: References: <501A23BE.8040706@donders.ru.nl> <4570E04B-810A-4002-B286-561FDE5C0221@donders.ru.nl> Message-ID: Dear Qi, Are you using the most current FieldTrip version? That line is 217 in the current version. But cfg.sphereradius should be set a few lines previous to that line, so I'm not sure how it is a non-existent field. What are the cfg options you use to call ft_sourceinterpolate? Best, Johanna 2012/8/2 qi li > Jan and Jörn: > > Thanks a lot for your answers! You did understand my questions pretty > well. I actually tried to use ft_sourceinterpolate to find the voxel > index without success. The error message is like > > 'Reference to non-existent field 'sphereradius'. > > Error in ft_sourceinterpolate (line 203) > interpmat = interp_ungridded(functional.pos, > warp_apply(anatomical.transform, [X(:) Y(:) Z(:)]), ...' > > On Thu, Aug 2, 2012 at 3:07 AM, jan-mathijs schoffelen > wrote: > > Hi Qi, > > > > To follow up on Jörn: > > > > If you reconstructed the source activations on a mesh that represents the > > cortical surface, you first need to interpolate the functional data into > a > > 3D volume, if you want to use ft_sourceplot for visualization. For this > you > > need ft_sourceinterpolate. > > As an alternative, to visualize the results, you can use ft_sourcemovie. > > However, this function is under development so it may not give the most > > visually appealing results. > > > > Best, > > > > JM > > > > > > On Aug 2, 2012, at 8:52 AM, Jörn M. Horschig wrote: > > > > Dear Qi, > > > > Not quite sure if I understand your request correctly, but you can just > > rehape: > > source.avg.pow = reshape(source.avg.pow, anatomical.dim); > > Then instead of being a vector, it will become a matrix. > > > > If that's not the answer you intended to get, please phrase your question > > differently :) > > > > Best, > > Jörn > > > > On 8/1/2012 7:33 PM, qi li wrote: > > > > Hi, > > > > Is there anyway to map the source space mesh back to voxel space so I > > can use the ft_sourceplot. > > > > After MNE source reconstruction, I have > > > > source = > > > > time: [1x399 double] > > pos: [8196x3 double] > > inside: [1x8196 double] > > outside: [1x0 double] > > method: 'average' > > avg: [1x1 struct] > > cfg: [1x1 struct] > > > > I have 8196 mesh points but how to map it back to the anatomical voxel > > space? Thanks! > > > > Qi > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > > > > -- > > Jörn M. Horschig > > PhD Student > > Donders Institute for Brain, Cognition and Behaviour > > Centre for Cognitive Neuroimaging > > Radboud University Nijmegen > > Neuronal Oscillations Group > > > > P.O. Box 9101 > > NL-6500 HB Nijmegen > > The Netherlands > > > > Contact: > > E-Mail: jm.horschig at donders.ru.nl > > Tel: +31-(0)24-36-68493 > > Web: http://www.ru.nl/donders > > > > Visiting address: > > Trigon, room 2.30 > > Kapittelweg 29 > > NL-6525 EN Nijmegen > > The Netherlands > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > > Jan-Mathijs Schoffelen, MD PhD > > > > Donders Institute for Brain, Cognition and Behaviour, > > Centre for Cognitive Neuroimaging, > > Radboud University Nijmegen, The Netherlands > > > > Max Planck Institute for Psycholinguistics, > > Nijmegen, The Netherlands > > > > J.Schoffelen at donders.ru.nl > > Telephone: +31-24-3614793 > > > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -------------- next part -------------- An HTML attachment was scrubbed... URL: From cs at imm.dtu.dk Tue Aug 7 14:40:46 2012 From: cs at imm.dtu.dk (Carsten Stahlhut) Date: Tue, 7 Aug 2012 14:40:46 +0200 Subject: [FieldTrip] Release of the SmartphoneBrainScanner2 Message-ID: [Apologies for cross-posting] Dear Fieldtrip list We are very happy to announce the release of the SmartphoneBrainScanner2 platform: http://code.google.com/p/smartphonebrainscanner2/ SmartphoneBrainScanner2 is a framework for building cross-platform real-time EEG applications. Originally developed at the Technical University of Denmark for collecting and analyzing signals from Emotiv EPOC headset, its extensible architecture allows working with various EEG systems and multiple platforms. *Cross Platform* SmartphoneBrainScanner2 is written in Qt, a C++ framework offering the power of the native development and unified support for multiple platforms. Plus the UI can be created in QML, high-level declarative UI framework. SBS2 can be compiled for every platform supporting Qt 4, including Linux, OSX, Windows, Android, Maemo 5, MeeGo. Although not yet attempted, it should also work on iOS and BlackBerry OS. *Advanced EEG* SmartphoneBrainScanner2 contains state-of-the-art techniques for working with multi-channel EEG signal in real-time, most notably source reconstruction methods with online adaptation to the noise level. Current implemented source reconstruction approaches cover the minimum-norm and low resolution tomography (LORETA) methods formulated in a Bayesian framework using a expectation-maximization scheme for hyperparameter estimation. The SBS2 source reconstruction is realized using a pre-build forward model connecting the cortical surface with the electrodes at the scalp. The current forward model provided with the software is a 3-spheres model obtained from the Matlab toolbox SPM8 using coarse spatial resolution and with sensor positions in accordance with the Emotiv EPOC system Besides, source reconstruction methods, additional machine learning methods such as independent component analysis (ICA), common spatial patterns (CSP), and Bayesian classifiers are continuously added. *New Approach* Real-time EEG doesn't have to happen in the lab! Consumer-grade and inexpensive research neuroheadsets allow for portability, delivering high-quality EEG signal. SmartphoneBrainScanner2 apps can be developed just like any other apps, featuring reach interface, connectivity, etc. Go, create! *How to cite SBS2* Please acknowledge the work of the SmartphoneBrainScanner2 by citing (Stopczynski et al, 2011). See also additional project related references at the homepage. A. Stopczynski, J. E. Larsen, C. Stahlhut, M. K. Petersen, & L. K. Hansen (2011), *A smartphone interface for a wireless EEG headset with real-time 3D reconstruction*, Affective Computing and Intelligent Interaction (ACII 2011), Lecture Notes in Computer Science (LNCS) 6357, Springer-Verlag Berlin Heidelberg, pp.317-318. *How to import data to Matlab* Demo scripts of how to import data to Matlab and to convert data into EEGLAB, FieldTrip, or SPM8 format are provided. Further description can be found at the Data Handling page: http://code.google.com/p/smartphonebrainscanner2/wiki/DataHandling?ts=1344340750&updated=DataHandling *Developer team* - Arkadiusz Stopczynski, DTU Informatics - Carsten Stahlhut, DTU Informatics - Michael Kai Petersen, DTU Informatics - Jakob Eg Larsen, DTU Informatics - Lars Kai Hansen, DTU Informatics *Acknowledgement* This work is supported in part by - Danish Lundbeck Foundation through Center for Integrated Molecular Brain Imaging (CIMBI ) - Danish Lundbeck Foundation through the project Real-time brain imaging by EEG - H C Ørsted Foundation - Nokia Best, Carsten -- Carsten Stahlhut Section for Cognitive Systems Department of Informatics and Mathematical Modelling Richard Petersens Plads, Building 321 Technical University of Denmark DK-2800 Kongens Lyngby, Denmark -------------- next part -------------- An HTML attachment was scrubbed... URL: From stan.vanpelt at fcdonders.ru.nl Wed Aug 8 16:14:38 2012 From: stan.vanpelt at fcdonders.ru.nl (Stan van Pelt) Date: Wed, 8 Aug 2012 16:14:38 +0200 (CEST) Subject: [FieldTrip] Fwd: source localization - MNI coordinates In-Reply-To: <1193658750.1278539.1344433666547.JavaMail.root@indus.zimbra.ru.nl> Message-ID: <1326953416.73376.1344435278179.JavaMail.root@sculptor.zimbra.ru.nl> Hi Alina, I forward your question to the FT discussion list, because I do not have experience with interpreting the source localization coordinate values. e.g., I do not know if they are actually in MNI space. ----- Doorgestuurd bericht ----- > Van: "A.S. Peter (Alina)" > Hi Stan, > > by visual inspection, I always thought that my source localization was > quite ok, but now I actually looked with MRICron where the spm > coordinates (=MNI coordinates, right?) are that fieldtrip outputs, and > some are near occipital regions but outside the brain... I am puzzled. > I tried linear ft_volumenormalize and also nonlinear, the MNI > coordinates are somewhat different but the problem remains. > > I then went back to look at SAM which we did first and it seems there > the sources were localized somewhat deeper/nicer than with LCMV > anyhow, which also tells me it's probably not a fiducial problem. I > think it may be partly my code (i.e. what I basically took from Avgis > and looks like yours) but also partly conversion to MNI causing > problems. Did you ever try to see the MNI coordinates for your data? > > Best, > > Alina -- Stan van Pelt Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen Kapittelweg 29, 6525 EN Nijmegen, Netherlands E-mail: stan.vanpelt at donders.ru.nl Website: www.ru.nl/donders/ Phone: (+31) (0)24 36 10981 Fax: (+31) (0)24 36 10989 From polomacnenad at gmail.com Thu Aug 9 11:56:13 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Thu, 9 Aug 2012 11:56:13 +0200 Subject: [FieldTrip] ordinal numbers of bad trials Message-ID: Hi, I would like to know is there an option to save ordinal number of bad trials after the artifact rejection step. I noticed that in all artifact rejection methods(ft_rejectvisual or ft_artifact_zvalue) I end up with structure "variable_name.cfg.artifact"which consist time points of the bad segments but I cannot find ordinal numbers of it. Kind regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From cornabel at googlemail.com Thu Aug 9 13:54:59 2012 From: cornabel at googlemail.com (cornelius abel) Date: Thu, 09 Aug 2012 13:54:59 +0200 Subject: [FieldTrip] ordinal numbers of bad trials In-Reply-To: References: Message-ID: <5023A513.7090208@googlemail.com> Hi Nenad, for the reject visual function i have a work around to this problem. First you have to add a column to the data.trialinfo table with a unique trial ID (in my case column 6). Then you can compare the output trialinfo of the rejectvisual function (dummy.trialinfo) to the original trialinfo and identify the rejected trials. In VisBadTrials you then have the unique trial ID of the rejected trials. Have a look at the code snip: ... dummy = ft_rejectvisual(cfg,data); VisBadTrials=data.trialinfo((find(~ismember(data.trialinfo(:,6),dummy.trialinfo(:,6)))),6)'; VisBadTrials=unique(VisBadTrials); Cornelius Am 09.08.2012 11:56, schrieb Nenad Polomac: > Hi, > > I would like to know is there an option to save ordinal number of bad > trials after the artifact rejection step. I noticed that in all > artifact rejection methods(ft_rejectvisual or ft_artifact_zvalue) I > end up with structure "variable_name.cfg.artifact"which consist time > points of the bad segments but I cannot find ordinal numbers of it. > > Kind regards! > > Nenad > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From Markus.Butz at uni-duesseldorf.de Thu Aug 9 17:24:48 2012 From: Markus.Butz at uni-duesseldorf.de (Markus Butz) Date: Thu, 09 Aug 2012 16:24:48 +0100 Subject: [FieldTrip] ft_rejectvisual: channel display Message-ID: <7570de0d8d01.5023e450@uni-duesseldorf.de> Dear list After defining 1 sec trials from continuous CTF data I want to use ft_rejectvisual using the following bit of code: cfg.method = 'trial'; chansel = ft_channelselection({'M*','EOG*','-B*','-P*','-R*','-Q*'}, data.label); cfg.channel = chansel; cfg.keepchannel = 'no'; data = ft_rejectvisual(cfg,data); What I don't understand is, why all channels are displayed in the layout, while only the selected ones show data and are saved in data.label (after going through >1 trials). However, when recalling "data = ft_rejectvisual(cfg,data);" only the selected ones are displayed in the layout with data. Does anyone has an explanation/fix for this, ie a solution that solely selected channels are displayed with the first call of ft_rejectvisual? Cheers Markus -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Fri Aug 10 14:43:23 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Fri, 10 Aug 2012 14:43:23 +0200 Subject: [FieldTrip] ordinal numbers of bad trials Message-ID: Dear Cornelius, Thank you very much for this cool solution! Kind regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From sauer.mpih at googlemail.com Fri Aug 10 15:49:47 2012 From: sauer.mpih at googlemail.com (Andreas Sauer) Date: Fri, 10 Aug 2012 15:49:47 +0200 Subject: [FieldTrip] ft_qualitycheck Message-ID: Dear all, I tried to check my MEG (CTF) datasets with ft_qualitycheck. Unfortunately it didn't work. An error occurs in the boxplot.m function in line 2020, where the data should be copied into the dataset. But for me, there is no dataset(). My question is whether this is due to my dataset-information or due to an error in the script? An how to solve this issue? Any help is highly appreciated! Thanks! Best regards, Andreas -- Andreas Sauer Max Planck Institute for Brain Research Department of Neurophysiology Deutschordenstraße 46 60528 Frankfurt am Main Germany T: +49 69 96769 278 F: +49 69 96769 327 Email: sauer.mpih at gmail.com www.brain.mpg.de -------------- next part -------------- An HTML attachment was scrubbed... URL: From a.stolk at fcdonders.ru.nl Fri Aug 10 17:17:06 2012 From: a.stolk at fcdonders.ru.nl (Stolk, A.) Date: Fri, 10 Aug 2012 17:17:06 +0200 (CEST) Subject: [FieldTrip] ft_qualitycheck In-Reply-To: Message-ID: <1868283926.90397.1344611826092.JavaMail.root@sculptor.zimbra.ru.nl> Hi Andreas, The boxplot function is a matlab statistics toolbox function. I do not see why it throws an error here. However, I did find a problem  related to the dataset history file (.hist) which contains details of the recording itself. I will fix this issue  right away. If you are still getting the same error  with tomorrow's version of FT, please give me shout. And preferably  provide some more details about the error, and how you are calling the function. best, Arjen ----- Oorspronkelijk bericht ----- > Van: "Andreas Sauer" > Aan: fieldtrip at science.ru.nl > Verzonden: Vrijdag 10 augustus 2012 15:49:47 > Onderwerp: [FieldTrip] ft_qualitycheck > Dear all, > I tried to check my MEG (CTF) datasets with ft_qualitycheck. > Unfortunately it didn't work. An error occurs in the boxplot.m > function in line 2020, where the data should be copied into the > dataset. But for me, there is no dataset(). > My question is whether this is due to my dataset-information or due to > an error in the script? An how to solve this issue? > Any help is highly appreciated! > Thanks! > Best regards, > Andreas > -- > Andreas Sauer > Max Planck Institute for Brain Research > Department of Neurophysiology > Deutschordenstraße 46 > 60528 Frankfurt am Main > Germany > T: +49 69 96769 278 > F: +49 69 96769 327 > Email: sauer.mpih at gmail.com > www.brain.mpg.de > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From sauer.mpih at googlemail.com Mon Aug 13 11:39:23 2012 From: sauer.mpih at googlemail.com (Andreas Sauer) Date: Mon, 13 Aug 2012 11:39:23 +0200 Subject: [FieldTrip] ft_qualitycheck In-Reply-To: <1868283926.90397.1344611826092.JavaMail.root@sculptor.zimbra.ru.nl> References: <1868283926.90397.1344611826092.JavaMail.root@sculptor.zimbra.ru.nl> Message-ID: Dear Arjen, thanks for your replay! I just tried with the current version of fieldtrip (20120812) but I still get this error message: ??? Undefined function or variable 'dataset'. Error in ==> boxplot>computeBoxLocation at 2020 boxValDs = dataset(); Error in ==> boxplot at 312 [boxVal,boxValPlot,boxIdx] = computeBoxLocation(xSorted,... Error in ==> ft_qualitycheck>draw_figure at 547 boxplot(h.HmotionAxes, hmotions, 'orientation', 'horizontal', 'notch', 'on'); Error in ==> ft_qualitycheck at 296 draw_figure(info, temp_timelock, temp_freq, temp_summary, temp_headpos, toi); I call the function ft_qualitycheck as described in the tutorial. cfg.dataset = subject_xy.ds; ft_qualitycheck(cfg). It runs the analysis, it writes the .mat-file but then it throws the error message. So to me, it seems that there is some information (the variable 'dataset') missing. But I don't know whether this is due to our recording or an error in one of the functions... Best, Andreas 2012/8/10 Stolk, A. > Hi Andreas, > > > > The boxplot function is a matlab statistics toolbox function. I do not see > why it throws an error here. However, I did find a problem related to the > dataset history file (.hist) which contains details of the recording > itself. I will fix this issue right away. If you are still getting the same > error with tomorrow's version of FT, please give me shout. And > preferably provide some more details about the error, and how you are > calling the function. > > > > best, > > Arjen > > > > > ------------------------------ > > *Van: *"Andreas Sauer" > *Aan: *fieldtrip at science.ru.nl > *Verzonden: *Vrijdag 10 augustus 2012 15:49:47 > *Onderwerp: *[FieldTrip] ft_qualitycheck > > > Dear all, > > I tried to check my MEG (CTF) datasets with ft_qualitycheck. Unfortunately > it didn't work. An error occurs in the boxplot.m function in line 2020, > where the data should be copied into the dataset. But for me, there is no > dataset(). > > My question is whether this is due to my dataset-information or due to an > error in the script? An how to solve this issue? > > Any help is highly appreciated! > > Thanks! > > Best regards, > > Andreas > > > -- > Andreas Sauer > Max Planck Institute for Brain Research > Department of Neurophysiology > Deutschordenstraße 46 > 60528 Frankfurt am Main > Germany > > T: +49 69 96769 278 > F: +49 69 96769 327 > Email: sauer.mpih at gmail.com > www.brain.mpg.de > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- Dipl.-Psych. Andreas Sauer Max Planck Institute for Brain Research Deutschordenstraße 46 60528 Frankfurt am Main Germany T: +49 69 96769 278 F: +49 69 96769 327 Email: sauer.mpih at gmail.com www.brain.mpg.de -------------- next part -------------- An HTML attachment was scrubbed... URL: From a.stolk at fcdonders.ru.nl Mon Aug 13 13:19:11 2012 From: a.stolk at fcdonders.ru.nl (Stolk, A.) Date: Mon, 13 Aug 2012 13:19:11 +0200 (CEST) Subject: [FieldTrip] ft_qualitycheck In-Reply-To: Message-ID: <1337301516.103006.1344856751003.JavaMail.root@sculptor.zimbra.ru.nl> Hi Andreas, Thanks for testing with the newest version. My suspicion that your bug pertains to the matlab function 'boxplot' has grown. Boxplot calls another function, 'dataset', that does exist in my version of matlab (2010b). To be more precise, it is located in the /toolbox/shared/statslib/@dataset/ directory. In my version of boxplot the line 'boxValDs = dataset();' is at line 2021 and not at line 2020. Please therefore check if your boxplot version, or your matlab version in general, is not corrupted by performing the following command: boxplot(1). Hope this helps, Arjen ----- Oorspronkelijk bericht ----- > Van: "Andreas Sauer" > Aan: "FieldTrip discussion list" > Verzonden: Maandag 13 augustus 2012 11:39:23 > Onderwerp: Re: [FieldTrip] ft_qualitycheck > Dear Arjen, > thanks for your replay! > I just tried with the current version of fieldtrip (20120812) but I > still get this error message: > ??? Undefined function or variable 'dataset'. > Error in ==> boxplot>computeBoxLocation at 2020 > boxValDs = dataset(); > Error in ==> boxplot at 312 > [boxVal,boxValPlot,boxIdx] = computeBoxLocation(xSorted,... > Error in ==> ft_qualitycheck>draw_figure at 547 >   boxplot(h.HmotionAxes, hmotions, 'orientation', 'horizontal', > 'notch', 'on'); > Error in ==> ft_qualitycheck at 296 >       draw_figure(info, temp_timelock, temp_freq, temp_summary, > temp_headpos, toi); > I call the function ft_qualitycheck as described in the tutorial. > cfg.dataset = subject_xy.ds; ft_qualitycheck(cfg). It runs the > analysis, it writes the .mat-file but then it throws the error > message. So to me, it seems that there is some information (the > variable 'dataset') missing. But I don't know whether this is due to > our recording or an error in one of the functions... > Best, > Andreas > 2012/8/10 Stolk, A. < a.stolk at fcdonders.ru.nl > > > Hi Andreas, > >   > > The boxplot function is a matlab statistics toolbox function. I do > > not > > see why it throws an error here. However, I did find a > > problem related > > to the dataset history file (.hist) which contains details of the > > recording itself. I will fix this issue right away. If you are still > > getting the same error with tomorrow's version of FT, please give me > > shout. And preferably provide some more details about the error, and > > how you are calling the function. > >   > > best, > > Arjen > >   > > > Van: "Andreas Sauer" < sauer.mpih at googlemail.com > > > > Aan: fieldtrip at science.ru.nl > > > Verzonden: Vrijdag 10 augustus 2012 15:49:47 > > > Onderwerp: [FieldTrip] ft_qualitycheck > > > Dear all, > > > I tried to check my MEG (CTF) datasets with ft_qualitycheck. > > > Unfortunately it didn't work. An error occurs in the boxplot.m > > > function in line 2020, where the data should be copied into the > > > dataset. But for me, there is no dataset(). > > > My question is whether this is due to my dataset-information or > > > due > > > to > > > an error in the script? An how to solve this issue? > > > Any help is highly appreciated! > > > Thanks! > > > Best regards, > > > Andreas > > > -- > > > Andreas Sauer > > > Max Planck Institute for Brain Research > > > Department of Neurophysiology > > > Deutschordenstraße 46 > > > 60528 Frankfurt am Main > > > Germany > > > T: +49 69 96769 278 > > > F: +49 69 96769 327 > > > Email: sauer.mpih at gmail.com > > > www.brain.mpg.de > > > _______________________________________________ > > > fieldtrip mailing list > > > fieldtrip at donders.ru.nl > > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- > Dipl.-Psych. Andreas Sauer > Max Planck Institute for Brain Research > Deutschordenstraße 46 > 60528 Frankfurt am Main > Germany > T: +49 69 96769 278 > F: +49 69 96769 327 > Email: sauer.mpih at gmail.com > www.brain.mpg.de > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From eelke.spaak at donders.ru.nl Mon Aug 13 13:30:36 2012 From: eelke.spaak at donders.ru.nl (Eelke Spaak) Date: Mon, 13 Aug 2012 13:30:36 +0200 Subject: [FieldTrip] ft_qualitycheck In-Reply-To: <1337301516.103006.1344856751003.JavaMail.root@sculptor.zimbra.ru.nl> References: <1337301516.103006.1344856751003.JavaMail.root@sculptor.zimbra.ru.nl> Message-ID: Hi Andreas, In addition to Arjen's comments, it might also be wise to execute 'which boxplot' on the Matlab command prompt, to see whether you have a version of boxplot on your path that is shadowing the stats toolbox version. Best, Eelke On 13 August 2012 13:19, Stolk, A. wrote: > Hi Andreas, > > > > Thanks for testing with the newest version. My suspicion that your bug > pertains to the matlab function 'boxplot' has grown. Boxplot calls another > function, 'dataset', that does exist in my version of matlab (2010b). To be > more precise, it is located in the /toolbox/shared/statslib/@dataset/ > directory. > > > > In my version of boxplot the line 'boxValDs = dataset();' is at line 2021 > and not at line 2020. Please therefore check if your boxplot version, or > your matlab version in general, is not corrupted by performing the following > command: boxplot(1). > > > > Hope this helps, > > Arjen > > ________________________________ > > Van: "Andreas Sauer" > Aan: "FieldTrip discussion list" > Verzonden: Maandag 13 augustus 2012 11:39:23 > Onderwerp: Re: [FieldTrip] ft_qualitycheck > > > Dear Arjen, > > thanks for your replay! > > I just tried with the current version of fieldtrip (20120812) but I still > get this error message: > > > ??? Undefined function or variable 'dataset'. > > Error in ==> boxplot>computeBoxLocation at 2020 > boxValDs = dataset(); > > Error in ==> boxplot at 312 > [boxVal,boxValPlot,boxIdx] = computeBoxLocation(xSorted,... > > Error in ==> ft_qualitycheck>draw_figure at 547 > boxplot(h.HmotionAxes, hmotions, 'orientation', 'horizontal', 'notch', > 'on'); > > Error in ==> ft_qualitycheck at 296 > draw_figure(info, temp_timelock, temp_freq, temp_summary, > temp_headpos, toi); > > > I call the function ft_qualitycheck as described in the tutorial. > cfg.dataset = subject_xy.ds; ft_qualitycheck(cfg). It runs the analysis, it > writes the .mat-file but then it throws the error message. So to me, it > seems that there is some information (the variable 'dataset') missing. But I > don't know whether this is due to our recording or an error in one of the > functions... > > Best, > > Andreas > > > > > 2012/8/10 Stolk, A. >> >> Hi Andreas, >> >> >> >> The boxplot function is a matlab statistics toolbox function. I do not see >> why it throws an error here. However, I did find a problem related to the >> dataset history file (.hist) which contains details of the recording itself. >> I will fix this issue right away. If you are still getting the same error >> with tomorrow's version of FT, please give me shout. And preferably provide >> some more details about the error, and how you are calling the function. >> >> >> >> best, >> >> Arjen >> >> >> >> >> ________________________________ >> >> Van: "Andreas Sauer" >> Aan: fieldtrip at science.ru.nl >> Verzonden: Vrijdag 10 augustus 2012 15:49:47 >> Onderwerp: [FieldTrip] ft_qualitycheck >> >> >> Dear all, >> >> I tried to check my MEG (CTF) datasets with ft_qualitycheck. Unfortunately >> it didn't work. An error occurs in the boxplot.m function in line 2020, >> where the data should be copied into the dataset. But for me, there is no >> dataset(). >> >> My question is whether this is due to my dataset-information or due to an >> error in the script? An how to solve this issue? >> >> Any help is highly appreciated! >> >> Thanks! >> >> Best regards, >> >> Andreas >> >> >> -- >> Andreas Sauer >> Max Planck Institute for Brain Research >> Department of Neurophysiology >> Deutschordenstraße 46 >> 60528 Frankfurt am Main >> Germany >> >> T: +49 69 96769 278 >> F: +49 69 96769 327 >> Email: sauer.mpih at gmail.com >> www.brain.mpg.de >> >> >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip >> >> >> >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > -- > Dipl.-Psych. Andreas Sauer > Max Planck Institute for Brain Research > Deutschordenstraße 46 > 60528 Frankfurt am Main > Germany > > T: +49 69 96769 278 > F: +49 69 96769 327 > Email: sauer.mpih at gmail.com > www.brain.mpg.de > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From maximilien.chaumon at gmail.com Mon Aug 13 14:16:56 2012 From: maximilien.chaumon at gmail.com (Maximilien Chaumon) Date: Mon, 13 Aug 2012 14:16:56 +0200 Subject: [FieldTrip] [Eeglablist] nested hypothesis testing to decide whether to use one or two dipoles to fit a component In-Reply-To: References: Message-ID: Hello eeglab & Fieldtrip, I'm trying to find out if it would be possible to use a nested hypothesis testing approach to decide whether to use a one or two dipole model while estimating components' dipole locations. The rationale I would like to follow is this: with two dipoles, we will always obtain a better fit than with one dipole, but the decrease in sum of squared errors (SSE) should follow a F distribution with k (= Nparameters_2dipoles - Nparameters_1dipole) degrees of freedom. If the decrease in SSE is greater than what would be expected under this F distribution, then we decide that 2 dipoles provide a sufficiently better fit and decide using them. I asked this question to eeglablist and Scott pointed out that it is difficult/impossible(?) to determine if the second dipole fits actual interesting data or just noise introduced by the imperfect head model. Christian then said it'd be worth a shot, and I agree, so here I am again with two questions, or two confirmations: 1) *How many parameters are estimated in ft_dipolefitting.m ?* specially in the case of 2 dipoles. If I count correctly, we estimate 6 parameters for one dipole, and, depending on whether the orientation has to be the same in the 2 dipoles, one (amplitude) or three (amplitude and orientation) more. 2) *Can I assimilate the relative residual variance to a SSE?* the function rv.m does this: rv = sum((d1-d2).^2) ./ sum(d1.^2); So that seems to be a sum of squared errors divided by the variance of the original data. So if I multiply the rv by the sum squared component map, I should get it, right? Thanks a lot! Max 2012/8/11 Christian Kothe > I can only speak from my armchair here, but it sounds like it should be > worth a try - even if you don't get the # of parameters exactly right it > will probably give you at least some level of complexity control in > whatever the range of validity is. If it works, it may inspire follow-up > work (e.g., Bayesian model selection or likelihood ratio tests). > > The number of parameters for a 2-dipole model seems to be 3 (xyz) + 4 (2x > the orientation parameters). Not sure about the momentum, though - you > might look up the place where the actual function minimization is being > performed in dipfit (fminunc call?) and see whether these are being > optimized together with the others. > > Christian > > On Aug 10, 2012, at 3:29 PM, Scott Makeig wrote: > > MAX - Unfortunately, in general using two dipoles rather than one will > ~always improve the fit. Even if the source is a pure single dipole, a > second dipole can be used to correct for noise or errors in the forward > head model. This is less always the case for the constrained > spatially-symmetric dipole pairs allowed by dipfit(). However, we have not > thought of an optimal way to decide between using one or (occasionally) two > dipoles to fit e.g. maps of ICA brain sources. The goal would be to decide > whether the two-dipole version is fitting noise/forward model error vs > actual bilateral source generation... > > Scott Makeig > > On Thu, Aug 9, 2012 at 1:54 AM, Maximilien Chaumon < > maximilien.chaumon at gmail.com> wrote: > >> Hello all, >> >> When fitting dipoles to components, we are all sooner or later puzzled by >> the question whether to use one or two symmetrical dipoles. >> >> Would it be correct to put the problems in terms of a nested hypothesis >> testing? >> >> We are fitting a scalp map with one or two parameters and get a residual >> variance after the fit. >> Could we not use this residual variance as a measure of the SSE and >> compute a F statistic to decide whether to use the more complex (with two >> dipoles) or simpler (with one dipole) of two nested models? >> If yes, then how would we decide on the number of degrees of freedom? How >> many free parameters do we have in each case? x,y,z,and two orientations >> per dipole? how does the imposed symmetry affect that number? Could we >> really map residual variance to SSE? How many "observations" do we have in >> that case (see formula below)? >> >> I found this formula, for F: >> F = (SSEF-SSER)/ (kF-kR) / ((1-SSEF)/(N-kF-1)) >> where >> SSE is sum of squared errors, >> k is numbers of parameters, >> N is number of observations (? what in our case?) >> F and R indices for full and reduced model respectively (in our case two >> and one dipole). >> >> >> Thanks a lot for any comment! >> Best, >> Max >> >> dipfit >> >> >> >> >> >> _______________________________________________ >> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html >> To unsubscribe, send an empty email to >> eeglablist-unsubscribe at sccn.ucsd.edu >> For digest mode, send an email with the subject "set digest mime" to >> eeglablist-request at sccn.ucsd.edu >> > > > > -- > Scott Makeig, Research Scientist and Director, Swartz Center for > Computational Neuroscience, Institute for Neural Computation; Prof. of > Neurosciences (Adj.), University of California San Diego, La Jolla CA > 92093-0559, http://sccn.ucsd.edu/~scott > > _______________________________________________ > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html > To unsubscribe, send an empty email to > eeglablist-unsubscribe at sccn.ucsd.edu > For digest mode, send an email with the subject "set digest mime" to > eeglablist-request at sccn.ucsd.edu > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From sauer.mpih at googlemail.com Mon Aug 13 14:59:09 2012 From: sauer.mpih at googlemail.com (Andreas Sauer) Date: Mon, 13 Aug 2012 14:59:09 +0200 Subject: [FieldTrip] ft_qualitycheck In-Reply-To: References: <1337301516.103006.1344856751003.JavaMail.root@sculptor.zimbra.ru.nl> Message-ID: Dear Eelke, dear Arjen, thanks for helping! Here, I analyze data on a cluster with Matlab 2008b. Although I found a folder "@dataset" in /toolbox/shared/statslib/, unfortunately there is no such function in it. Since I found the function in my local Matlab, which is version 2010a, I guess that this is a feature of newer Matlab versions. I downloaded one dataset to my local machine and with Matlab 2010a it worked... I am not sure whether it might be good to note that in the tutorial... Best, Andreas 2012/8/13 Eelke Spaak > Hi Andreas, > > In addition to Arjen's comments, it might also be wise to execute > 'which boxplot' on the Matlab command prompt, to see whether you have > a version of boxplot on your path that is shadowing the stats toolbox > version. > > Best, > Eelke > > On 13 August 2012 13:19, Stolk, A. wrote: > > Hi Andreas, > > > > > > > > Thanks for testing with the newest version. My suspicion that your bug > > pertains to the matlab function 'boxplot' has grown. Boxplot calls > another > > function, 'dataset', that does exist in my version of matlab (2010b). To > be > > more precise, it is located in the /toolbox/shared/statslib/@dataset/ > > directory. > > > > > > > > In my version of boxplot the line 'boxValDs = dataset();' is at line 2021 > > and not at line 2020. Please therefore check if your boxplot version, or > > your matlab version in general, is not corrupted by performing the > following > > command: boxplot(1). > > > > > > > > Hope this helps, > > > > Arjen > > > > ________________________________ > > > > Van: "Andreas Sauer" > > Aan: "FieldTrip discussion list" > > Verzonden: Maandag 13 augustus 2012 11:39:23 > > Onderwerp: Re: [FieldTrip] ft_qualitycheck > > > > > > Dear Arjen, > > > > thanks for your replay! > > > > I just tried with the current version of fieldtrip (20120812) but I still > > get this error message: > > > > > > ??? Undefined function or variable 'dataset'. > > > > Error in ==> boxplot>computeBoxLocation at 2020 > > boxValDs = dataset(); > > > > Error in ==> boxplot at 312 > > [boxVal,boxValPlot,boxIdx] = computeBoxLocation(xSorted,... > > > > Error in ==> ft_qualitycheck>draw_figure at 547 > > boxplot(h.HmotionAxes, hmotions, 'orientation', 'horizontal', 'notch', > > 'on'); > > > > Error in ==> ft_qualitycheck at 296 > > draw_figure(info, temp_timelock, temp_freq, temp_summary, > > temp_headpos, toi); > > > > > > I call the function ft_qualitycheck as described in the tutorial. > > cfg.dataset = subject_xy.ds; ft_qualitycheck(cfg). It runs the analysis, > it > > writes the .mat-file but then it throws the error message. So to me, it > > seems that there is some information (the variable 'dataset') missing. > But I > > don't know whether this is due to our recording or an error in one of the > > functions... > > > > Best, > > > > Andreas > > > > > > > > > > 2012/8/10 Stolk, A. > >> > >> Hi Andreas, > >> > >> > >> > >> The boxplot function is a matlab statistics toolbox function. I do not > see > >> why it throws an error here. However, I did find a problem related to > the > >> dataset history file (.hist) which contains details of the recording > itself. > >> I will fix this issue right away. If you are still getting the same > error > >> with tomorrow's version of FT, please give me shout. And preferably > provide > >> some more details about the error, and how you are calling the function. > >> > >> > >> > >> best, > >> > >> Arjen > >> > >> > >> > >> > >> ________________________________ > >> > >> Van: "Andreas Sauer" > >> Aan: fieldtrip at science.ru.nl > >> Verzonden: Vrijdag 10 augustus 2012 15:49:47 > >> Onderwerp: [FieldTrip] ft_qualitycheck > >> > >> > >> Dear all, > >> > >> I tried to check my MEG (CTF) datasets with ft_qualitycheck. > Unfortunately > >> it didn't work. An error occurs in the boxplot.m function in line 2020, > >> where the data should be copied into the dataset. But for me, there is > no > >> dataset(). > >> > >> My question is whether this is due to my dataset-information or due to > an > >> error in the script? An how to solve this issue? > >> > >> Any help is highly appreciated! > >> > >> Thanks! > >> > >> Best regards, > >> > >> Andreas > >> > >> > >> -- > >> Andreas Sauer > >> Max Planck Institute for Brain Research > >> Department of Neurophysiology > >> Deutschordenstraße 46 > >> 60528 Frankfurt am Main > >> Germany > >> > >> T: +49 69 96769 278 > >> F: +49 69 96769 327 > >> Email: sauer.mpih at gmail.com > >> www.brain.mpg.de > >> > >> > >> _______________________________________________ > >> fieldtrip mailing list > >> fieldtrip at donders.ru.nl > >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > >> > >> > >> > >> _______________________________________________ > >> fieldtrip mailing list > >> fieldtrip at donders.ru.nl > >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > > > > > > -- > > Dipl.-Psych. Andreas Sauer > > Max Planck Institute for Brain Research > > Deutschordenstraße 46 > > 60528 Frankfurt am Main > > Germany > > > > T: +49 69 96769 278 > > F: +49 69 96769 327 > > Email: sauer.mpih at gmail.com > > www.brain.mpg.de > > > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- Dipl.-Psych. Andreas Sauer Max Planck Institute for Brain Research Deutschordenstraße 46 60528 Frankfurt am Main Germany T: +49 69 96769 278 F: +49 69 96769 327 Email: sauer.mpih at gmail.com www.brain.mpg.de -------------- next part -------------- An HTML attachment was scrubbed... URL: From akiko.ikkai at gmail.com Tue Aug 14 22:19:13 2012 From: akiko.ikkai at gmail.com (Akiko Ikkai) Date: Tue, 14 Aug 2012 16:19:13 -0400 Subject: [FieldTrip] actvsblT and average over time and/or frequency Message-ID: Hi all, I'm testing for the power difference between baseline (fixation) and memory delay periods, following the examples in "Permutation Tests > Within trial experiments" tutorial. When I set cfg.statistic = 'actvsblT'; and cfg.avgovertime = 'yes'; cfg.avgoverfreq = 'yes'; then run ft_freqstatistics, I get an error in statfun_actvsblT.m, "Inappropriate dimord for the statistics function STATFUN_ACTVSBLT.". This is because with avgovertime and/or avgoverfreq set to 'yes', cfg.dimord for data input to statfun_actvsblT is simply "chan," whereas statfun_actvsblT requires "chan_freq_time." When I comment out cfg.avgovertime AND freq, permutation runs fine. This is particular to statfun_actvsblT; other functions, such as statfun_depsamplesT, do not have this dimord structure requirements. I'm puzzled why averaging over time and/or frequency could not be a valid option for actvsblT... could someone help me understand? Thanks! Akiko -- Akiko Ikkai, Ph.D. Postdoctoral Fellow Department of Psychological and Brain Sciences Johns Hopkins University Ames Hall, 3400 N. Charles St. Baltimore, MD 21218 -------------- next part -------------- An HTML attachment was scrubbed... URL: From aler1 at gmx.de Wed Aug 15 15:47:20 2012 From: aler1 at gmx.de (alla Brodski) Date: Wed, 15 Aug 2012 15:47:20 +0200 Subject: [FieldTrip] Source analysis power unit Message-ID: <20120815134720.43010@gmx.net> Dear fieldtrip users, I am using fieldtrip for beamforming source analysis. I try to interpret the units of the source power results, which are for my data in the range of 10^-07 to 10^-10. I am using the fieldtrip version of 01.05.2012; with an older version (e.g. of 2009) I get the same numbers but in a range of 10^27 to 10^30. I have already noticed that there was a change for the singleshell method, which I am using; a new scaling factor was introduced in the beginning of 2011: http://mailman.science.ru.nl/pipermail/fieldtrip/2011-January/003367.html However, I am still wondering whether I can interpret the results in physical units or not. Can they still be interpreted in tesla /fT (as if they had the old scale)? Or am I possibly doing something wrong as 10^-07 to 10^-10 does not seem a meaningful range. Thank you very much! Alla Brodski From aler1 at gmx.de Wed Aug 15 16:07:16 2012 From: aler1 at gmx.de (alla Brodski) Date: Wed, 15 Aug 2012 16:07:16 +0200 Subject: [FieldTrip] Source analysis power unit In-Reply-To: <20120815134720.43010@gmx.net> References: <20120815134720.43010@gmx.net> Message-ID: <20120815140716.43010@gmx.net> Sorry, of course it is 10^-27 to 10^-30 with an old version -------- Original-Nachricht -------- > Datum: Wed, 15 Aug 2012 15:47:20 +0200 > Von: "alla Brodski" > An: fieldtrip at science.ru.nl > Betreff: [FieldTrip] Source analysis power unit > Dear fieldtrip users, > > I am using fieldtrip for beamforming source analysis. > I try to interpret the units of the source power results, which are for my > data in the range of 10^-07 to 10^-10. I am using the fieldtrip version of > 01.05.2012; with an older version (e.g. of 2009) I get the same numbers > but in a range of 10^27 to 10^30. > I have already noticed that there was a change for the singleshell method, > which I am using; a new scaling factor was introduced in the beginning of > 2011: > http://mailman.science.ru.nl/pipermail/fieldtrip/2011-January/003367.html > > However, I am still wondering whether I can interpret the results in > physical units or not. Can they still be interpreted in tesla /fT (as if they > had the old scale)? Or am I possibly doing something wrong as 10^-07 to > 10^-10 does not seem a meaningful range. > > Thank you very much! > > Alla Brodski > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From Paul_C at gmx.de Fri Aug 17 01:12:04 2012 From: Paul_C at gmx.de (Paul Kertscher) Date: Fri, 17 Aug 2012 01:12:04 +0200 Subject: [FieldTrip] Question about ft_dipolesimulation Message-ID: <20120816231204.162220@gmx.net> Dear fieldtrippers, I recetly asked myself what units the signal in ft_dipolsimulation had. First I thought, that it had to be a current, thus Ampére. After rechecking it in the Kybic et al. paper it obviously is an current density. If the signal simulated has an amplitude of 1, does this mean, that the current density would be 1A/m², or are these units in the function rather of an arbitrary kind? Best regards, Paul Kertscher --- Paul Kertscher (Dipl.-Phys.) Björnsonstr. 25 12163 Berlin +49/(0)30/221609359 +49/(0)1578/7839933 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Andrew.Dimitrijevic at cchmc.org Fri Aug 17 04:41:40 2012 From: Andrew.Dimitrijevic at cchmc.org (Dimitrijevic, Andrew) Date: Fri, 17 Aug 2012 02:41:40 +0000 Subject: [FieldTrip] postdoc position available ... auditory ERPs in cochlear implant users Message-ID: <1787E0F7D5DAD54E803CBB4E877A33BE026B7C90@MCEXMB3.chmccorp.cchmc.org> Dear FieldTrippers ... postdoc position available, feel free to pass along (sorry for cross posting) Postdoctoral position in auditory EPs with cochlear implant users We are seeking a highly motivated postdoctoral fellow with expertise in EEGLAB/FieldTrip to the join the Hearing Lab at the Communication Sciences Research Center (CSRC) at Cincinnati Children’s Hospital Medical Center (CCHMC). This particular project will involve psychoacoustics and high-density EEG recordings in adults and children with cochlear implants. Much of the work will involve post processing in Matlab therefore solid programming ability is essential. The ideal candidate will have a PhD in cognitive science/neuroscience/computer science/psychology/engineering or related discipline, and an interest in cochlear implants and EEG. The CSRC has a strong interdisciplinary community that has both clinicians and basic scientists. Our facilities include an EEG lab, a Neuroimaging center with research dedicated 3T MRI scanner, 275-channel MEG lab. The position is for 2 years and second year is contingent on satisfactory progress. The anticipated start date is Jan 2012. Applicants should submit a cover letter and CV, including the names and contact information of three references. Please send application materials by e-mail to: Andrew Dimitrijevic PhD Assistant Professor Communication Sciences Research Center https://csrc.cchmc.org/ andrew.dimitrijevic at cchmc.org https://csrc.cchmc.org/andrew-dimitrijevic CCHMC has a longstanding commitment to achieving diversity among faculty, staff, and students. CCHMC is an EO/AA Employer. -------------- next part -------------- An HTML attachment was scrubbed... URL: From julian.keil at gmail.com Fri Aug 17 14:50:25 2012 From: julian.keil at gmail.com (Julian Keil) Date: Fri, 17 Aug 2012 14:50:25 +0200 Subject: [FieldTrip] FIeldTrip Tutorian Videos Message-ID: <82B32158-73E9-4E12-B91A-C398B996EB34@gmail.com> Dear FieldTrippers, a while ago I held a little workshop at the BRAMS Institute in Montreal covering the basic steps in FieldTrip. I recorded the whole session and finally got around to put the videos online. Please feel free to check it out at or point any new users to it: http://vimeo.com/user11934546/videos The accompanying Matlab script can be found here: http://pastebin.com/yskmVKAh Please also tell me if I made any errors or you feel that I have explained something incorrectly. Best, Julian ******************************************** Dr. Julian Keil International Laboratory for Brain Music and Sound Research (BRAMS) Pavillon 1430 Mont-Royal Université de Montréal Montréal, Québec Canada, H2V 4P3 julian.keil at gmail.com +1-514-343-6111-29653 www.brams.org -------------- next part -------------- An HTML attachment was scrubbed... URL: From pgoodin at swin.edu.au Sun Aug 19 13:17:16 2012 From: pgoodin at swin.edu.au (Peter Goodin) Date: Sun, 19 Aug 2012 11:17:16 +0000 Subject: [FieldTrip] Wavelet advice Message-ID: Hi Fieldtrip list, I'm attempting to use wavelets to analyse some data collected with a neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is 1.6 seconds) but am new to time frequency analysis. I'm interested in lower frequency bins (~4 to 20 Hz). The config settings I've been playing with are as follows: cfg.method = 'wavelet'; cfg.channel = 'MEG'; cfg.keeptrials = 'yes'; cfg.foi = [4:1:20] cfg.toi = [-.1:.01:1.5] cfg.width = x Using a default width of 7, I'm getting large boundary effects leading to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. I've played with the width and have found that a width of 1 gives only a small amount of boundary effect at the extreme edges and shows a large increase in the frequencies I'm interested in at a time point which corresponds quite nicely with the ERF data also analysed. Having read the Tallon-Baudry (1999) article, material from the tutorials and items from the mailing list, I understand that by using a wavelet width of 1 I've biased my results towards higher temporal vs. spectral resolution, but considering the low range of frequencies I'm interest in, is this a problem? Does using a low width with my cfg.foi set as it is just lead to a large amount of redundant data? Advice on this subject would be greatly appreciated. Peter. -------------- next part -------------- An HTML attachment was scrubbed... URL: From smoratti at psi.ucm.es Sun Aug 19 14:19:26 2012 From: smoratti at psi.ucm.es (smoratti at psi.ucm.es) Date: Sun, 19 Aug 2012 14:19:26 +0200 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: Message-ID: Dear Peter, I think you should use 5 cycles as minimum to get stable results. One cycle is not much to estimate the power of your time-freq of interest. Best, Stephan ________________________________________________________ Stephan Moratti, PhD see also: http://web.me.com/smoratti/ Universidad Complutense de Madrid Facultad de Psicología Departamento de Psicología Básica I Campus de Somosaguas 28223 Pozuelo de Alarcón (Madrid) Spain and Center for Biomedical Technology Laboratory for Cognitive and Computational Neuroscience Parque Científico y Tecnológico de la Universidad Politecnica de Madrid Campus Montegancedo 28223 Pozuelo de Alarcón (Madrid) Spain email: smoratti at psi.ucm.es Tel.: +34 679219982 El 19/08/2012, a las 13:17, Peter Goodin escribió: > Hi Fieldtrip list, > > I'm attempting to use wavelets to analyse some data collected with a neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is 1.6 seconds) but am new to time frequency analysis. I'm interested in lower frequency bins (~4 to 20 Hz). The config settings I've been playing with are as follows: > > cfg.method = 'wavelet'; > cfg.channel = 'MEG'; > cfg.keeptrials = 'yes'; > cfg.foi = [4:1:20] > cfg.toi = [-.1:.01:1.5] > cfg.width = x > > Using a default width of 7, I'm getting large boundary effects leading to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. I've played with the width and have found that a width of 1 gives only a small amount of boundary effect at the extreme edges and shows a large increase in the frequencies I'm interested in at a time point which corresponds quite nicely with the ERF data also analysed. > > Having read the Tallon-Baudry (1999) article, material from the tutorials and items from the mailing list, I understand that by using a wavelet width of 1 I've biased my results towards higher temporal vs. spectral resolution, but considering the low range of frequencies I'm interest in, is this a problem? Does using a low width with my cfg.foi set as it is just lead to a large amount of redundant data? > > Advice on this subject would be greatly appreciated. > > Peter. > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From sherrykhan78 at gmail.com Sun Aug 19 22:58:21 2012 From: sherrykhan78 at gmail.com (Sheraz Khan) Date: Sun, 19 Aug 2012 16:58:21 -0400 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: Message-ID: Hi, You can always try variables number of cycles may be start with 3 and then go to 5. Sheraz Martinos Center MGH/MIT/Harvard On Aug 19, 2012 7:17 AM, "Peter Goodin" wrote: > Hi Fieldtrip list, > > I'm attempting to use wavelets to analyse some data collected with a > neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is > 1.6 seconds) but am new to time frequency analysis. I'm interested in lower > frequency bins (~4 to 20 Hz). The config settings I've been playing with > are as follows: > > cfg.method = 'wavelet'; > cfg.channel = 'MEG'; > cfg.keeptrials = 'yes'; > cfg.foi = [4:1:20] > cfg.toi = [-.1:.01:1.5] > cfg.width = x > > Using a default width of 7, I'm getting large boundary effects leading > to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. > I've played with the width and have found that a width of 1 gives only a > small amount of boundary effect at the extreme edges and shows a large > increase in the frequencies I'm interested in at a time point which > corresponds quite nicely with the ERF data also analysed. > > Having read the Tallon-Baudry (1999) article, material from the > tutorials and items from the mailing list, I understand that by using a > wavelet width of 1 I've biased my results towards higher temporal vs. > spectral resolution, but considering the low range of frequencies I'm > interest in, is this a problem? Does using a low width with my cfg.foi set > as it is just lead to a large amount of redundant data? > > Advice on this subject would be greatly appreciated. > > Peter. > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pgoodin at swin.edu.au Mon Aug 20 01:20:49 2012 From: pgoodin at swin.edu.au (Peter Goodin) Date: Sun, 19 Aug 2012 23:20:49 +0000 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: , Message-ID: Thanks for all the replies so far. Re-reading my previous question I appear to have been a bit vague to the overall problem. The problem is I've used wavelets widths starting from 1 and increased to the default 7, but only a value of 1 gives me any usable data around 4Hz. What confuses me is that Kaplan et al, 2011 (http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001267) were able to get theta band activity using epochs of 1s (200ms pre 800ms post) without apparent boundary effects at a width of 5, but when I tried the same thing (including zero padding), the results were not usable. Again, any advice and explanations as to why 1. a value of 1 is not usable and 2. why I may be getting these boundary effects would be greatly appreciated. Peter ________________________________ From: fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl] on behalf of Sheraz Khan [sherrykhan78 at gmail.com] Sent: Monday, 20 August 2012 6:58 AM To: FieldTrip discussion list Subject: Re: [FieldTrip] Wavelet advice Hi, You can always try variables number of cycles may be start with 3 and then go to 5. Sheraz Martinos Center MGH/MIT/Harvard On Aug 19, 2012 7:17 AM, "Peter Goodin" > wrote: Hi Fieldtrip list, I'm attempting to use wavelets to analyse some data collected with a neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is 1.6 seconds) but am new to time frequency analysis. I'm interested in lower frequency bins (~4 to 20 Hz). The config settings I've been playing with are as follows: cfg.method = 'wavelet'; cfg.channel = 'MEG'; cfg.keeptrials = 'yes'; cfg.foi = [4:1:20] cfg.toi = [-.1:.01:1.5] cfg.width = x Using a default width of 7, I'm getting large boundary effects leading to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. I've played with the width and have found that a width of 1 gives only a small amount of boundary effect at the extreme edges and shows a large increase in the frequencies I'm interested in at a time point which corresponds quite nicely with the ERF data also analysed. Having read the Tallon-Baudry (1999) article, material from the tutorials and items from the mailing list, I understand that by using a wavelet width of 1 I've biased my results towards higher temporal vs. spectral resolution, but considering the low range of frequencies I'm interest in, is this a problem? Does using a low width with my cfg.foi set as it is just lead to a large amount of redundant data? Advice on this subject would be greatly appreciated. Peter. _______________________________________________ fieldtrip mailing list fieldtrip at donders.ru.nl http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From sherrykhan78 at gmail.com Mon Aug 20 03:29:42 2012 From: sherrykhan78 at gmail.com (Sheraz Khan) Date: Sun, 19 Aug 2012 21:29:42 -0400 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: Message-ID: You can not avoid the border effects they comes from the convolution. Only possible solution which I sometimes use when the epochs are shorter due to the design, is to do the wavelet decomposition on the continuous data, and then epoch it, this will avoid the borders from the epoch. Sheraz Martinos Center MGH/MIT/Harvard On Sun, Aug 19, 2012 at 7:20 PM, Peter Goodin wrote: > Thanks for all the replies so far. > > Re-reading my previous question I appear to have been a bit vague to the > overall problem. > > The problem is I've used wavelets widths starting from 1 and increased > to the default 7, but only a value of 1 gives me any usable data around > 4Hz. > > What confuses me is that Kaplan et al, 2011 ( > http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001267) > were able to get theta band activity using epochs of 1s (200ms pre 800ms > post) without apparent boundary effects at a width of 5, but when I tried > the same thing (including zero padding), the results were not usable. > > Again, any advice and explanations as to why 1. a value of 1 is not > usable and 2. why I may be getting these boundary effects would be greatly > appreciated. > > Peter > ------------------------------ > *From:* fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl] > on behalf of Sheraz Khan [sherrykhan78 at gmail.com] > *Sent:* Monday, 20 August 2012 6:58 AM > *To:* FieldTrip discussion list > *Subject:* Re: [FieldTrip] Wavelet advice > > Hi, > You can always try variables number of cycles may be start with 3 and then > go to 5. > Sheraz > Martinos Center > MGH/MIT/Harvard > On Aug 19, 2012 7:17 AM, "Peter Goodin" wrote: > >> Hi Fieldtrip list, >> >> I'm attempting to use wavelets to analyse some data collected with a >> neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is >> 1.6 seconds) but am new to time frequency analysis. I'm interested in lower >> frequency bins (~4 to 20 Hz). The config settings I've been playing with >> are as follows: >> >> cfg.method = 'wavelet'; >> cfg.channel = 'MEG'; >> cfg.keeptrials = 'yes'; >> cfg.foi = [4:1:20] >> cfg.toi = [-.1:.01:1.5] >> cfg.width = x >> >> Using a default width of 7, I'm getting large boundary effects leading >> to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. >> I've played with the width and have found that a width of 1 gives only a >> small amount of boundary effect at the extreme edges and shows a large >> increase in the frequencies I'm interested in at a time point which >> corresponds quite nicely with the ERF data also analysed. >> >> Having read the Tallon-Baudry (1999) article, material from the >> tutorials and items from the mailing list, I understand that by using a >> wavelet width of 1 I've biased my results towards higher temporal vs. >> spectral resolution, but considering the low range of frequencies I'm >> interest in, is this a problem? Does using a low width with my cfg.foi set >> as it is just lead to a large amount of redundant data? >> >> Advice on this subject would be greatly appreciated. >> >> Peter. >> >> >> >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip >> > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pgoodin at swin.edu.au Mon Aug 20 05:18:09 2012 From: pgoodin at swin.edu.au (Peter Goodin) Date: Mon, 20 Aug 2012 03:18:09 +0000 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: , Message-ID: Thanks for that! Much appreciated. ________________________________ From: fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl] on behalf of Sheraz Khan [sherrykhan78 at gmail.com] Sent: Monday, 20 August 2012 11:29 AM To: FieldTrip discussion list Subject: Re: [FieldTrip] Wavelet advice You can not avoid the border effects they comes from the convolution. Only possible solution which I sometimes use when the epochs are shorter due to the design, is to do the wavelet decomposition on the continuous data, and then epoch it, this will avoid the borders from the epoch. Sheraz Martinos Center MGH/MIT/Harvard On Sun, Aug 19, 2012 at 7:20 PM, Peter Goodin > wrote: Thanks for all the replies so far. Re-reading my previous question I appear to have been a bit vague to the overall problem. The problem is I've used wavelets widths starting from 1 and increased to the default 7, but only a value of 1 gives me any usable data around 4Hz. What confuses me is that Kaplan et al, 2011 (http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001267) were able to get theta band activity using epochs of 1s (200ms pre 800ms post) without apparent boundary effects at a width of 5, but when I tried the same thing (including zero padding), the results were not usable. Again, any advice and explanations as to why 1. a value of 1 is not usable and 2. why I may be getting these boundary effects would be greatly appreciated. Peter ________________________________ From: fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl] on behalf of Sheraz Khan [sherrykhan78 at gmail.com] Sent: Monday, 20 August 2012 6:58 AM To: FieldTrip discussion list Subject: Re: [FieldTrip] Wavelet advice Hi, You can always try variables number of cycles may be start with 3 and then go to 5. Sheraz Martinos Center MGH/MIT/Harvard On Aug 19, 2012 7:17 AM, "Peter Goodin" > wrote: Hi Fieldtrip list, I'm attempting to use wavelets to analyse some data collected with a neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is 1.6 seconds) but am new to time frequency analysis. I'm interested in lower frequency bins (~4 to 20 Hz). The config settings I've been playing with are as follows: cfg.method = 'wavelet'; cfg.channel = 'MEG'; cfg.keeptrials = 'yes'; cfg.foi = [4:1:20] cfg.toi = [-.1:.01:1.5] cfg.width = x Using a default width of 7, I'm getting large boundary effects leading to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. I've played with the width and have found that a width of 1 gives only a small amount of boundary effect at the extreme edges and shows a large increase in the frequencies I'm interested in at a time point which corresponds quite nicely with the ERF data also analysed. Having read the Tallon-Baudry (1999) article, material from the tutorials and items from the mailing list, I understand that by using a wavelet width of 1 I've biased my results towards higher temporal vs. spectral resolution, but considering the low range of frequencies I'm interest in, is this a problem? Does using a low width with my cfg.foi set as it is just lead to a large amount of redundant data? Advice on this subject would be greatly appreciated. Peter. _______________________________________________ fieldtrip mailing list fieldtrip at donders.ru.nl http://mailman.science.ru.nl/mailman/listinfo/fieldtrip _______________________________________________ fieldtrip mailing list fieldtrip at donders.ru.nl http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From nuria.donamayor at neuro.uni-luebeck.de Tue Aug 21 12:50:56 2012 From: nuria.donamayor at neuro.uni-luebeck.de (=?iso-8859-1?Q?Nuria_Do=F1amayor_Alonso?=) Date: Tue, 21 Aug 2012 12:50:56 +0200 Subject: [FieldTrip] source reconstruction on emfs Message-ID: <810A8E06C75EB447A8CEB73DBFD7BB0E4C03CD65AD@solaris.neuro.uni-luebeck.de> Dear fieldtrippers, I am currently analyzing a study, measured with a Neuromag 306 system, in which we stimulated the median nerve. I had never done source reconstruction with fieldtrip, so I followed the tutorial using MNE for EMF source analysis (which worked fine), only to find out that I could neither average individual sources nor perform stats, which is exactly what I need to do... So the question is: What would you guys recommend? Should I just try using the LCMV beamfomer? However, that might not be the best solution according to this thread http://mailman.science.ru.nl/pipermail/fieldtrip/2011-September/004321.html... Is there any way I could efficiently export my preprocessed/averaged data to some other software (MNE/SPM/Brainstorm...) and compute my source reconstruction and stats there? What do you guys usually do? Thanks a lot for your help, Nuria From jm.horschig at donders.ru.nl Tue Aug 21 13:24:21 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Tue, 21 Aug 2012 13:24:21 +0200 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: , Message-ID: <50336FE5.30602@donders.ru.nl> Hey Peter, when doing what Sheraz suggested you should, however, keep in mind that the edges are computed using data outside your epoch, i.e. possible muscle artifacts and other stuff (e.g. stimulation) will corrupt the frequency estimation if not properly accounted for. Make sure to note that when interpreting your data. Best, Jörn On 8/20/2012 5:18 AM, Peter Goodin wrote: > Thanks for that! Much appreciated. > ------------------------------------------------------------------------ > *From:* fieldtrip-bounces at science.ru.nl > [fieldtrip-bounces at science.ru.nl] on behalf of Sheraz Khan > [sherrykhan78 at gmail.com] > *Sent:* Monday, 20 August 2012 11:29 AM > *To:* FieldTrip discussion list > *Subject:* Re: [FieldTrip] Wavelet advice > > You can not avoid the border effects they comes from the convolution. > > Only possible solution which I sometimes use when the epochs are > shorter due to the design, is to do the wavelet decomposition on > the continuous data, and then epoch it, this will avoid the borders > from the epoch. > > Sheraz > Martinos Center > MGH/MIT/Harvard > > On Sun, Aug 19, 2012 at 7:20 PM, Peter Goodin > wrote: > > Thanks for all the replies so far. > > Re-reading my previous question I appear to have been a bit vague > to the overall problem. > > The problem is I've used wavelets widths starting from 1 and > increased to the default 7, but only a value of 1 gives me any > usable data around 4Hz. > > What confuses me is that Kaplan et al, 2011 > (http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001267) > were able to get theta band activity using epochs of 1s (200ms pre > 800ms post) without apparent boundary effects at a width of 5, but > when I tried the same thing (including zero padding), the results > were not usable. > > Again, any advice and explanations as to why 1. a value of 1 is > not usable and 2. why I may be getting these boundary effects > would be greatly appreciated. > > Peter > ------------------------------------------------------------------------ > *From:* fieldtrip-bounces at science.ru.nl > > [fieldtrip-bounces at science.ru.nl > ] on behalf of Sheraz Khan > [sherrykhan78 at gmail.com ] > *Sent:* Monday, 20 August 2012 6:58 AM > *To:* FieldTrip discussion list > *Subject:* Re: [FieldTrip] Wavelet advice > > Hi, > You can always try variables number of cycles may be start with 3 > and then go to 5. > Sheraz > Martinos Center > MGH/MIT/Harvard > > On Aug 19, 2012 7:17 AM, "Peter Goodin" > wrote: > > Hi Fieldtrip list, > > I'm attempting to use wavelets to analyse some data collected > with a neuromag system (sample rate of 500Hz, pre-trigger > period is 200ms, post is 1.6 seconds) but am new to time > frequency analysis. I'm interested in lower frequency bins (~4 > to 20 Hz). The config settings I've been playing with are as > follows: > > cfg.method = 'wavelet'; > cfg.channel = 'MEG'; > cfg.keeptrials = 'yes'; > cfg.foi = [4:1:20] > cfg.toi = [-.1:.01:1.5] > cfg.width = x > > Using a default width of 7, I'm getting large boundary effects > leading to a period of ~300ms (400 - 700ms post trigger) of > calculated data at 4Hz. I've played with the width and have > found that a width of 1 gives only a small amount of boundary > effect at the extreme edges and shows a large increase in the > frequencies I'm interested in at a time point which > corresponds quite nicely with the ERF data also analysed. > > Having read the Tallon-Baudry (1999) article, material from > the tutorials and items from the mailing list, I understand > that by using a wavelet width of 1 I've biased my results > towards higher temporal vs. spectral resolution, but > considering the low range of frequencies I'm interest in, is > this a problem? Does using a low width with my cfg.foi set as > it is just lead to a large amount of redundant data? > > Advice on this subject would be greatly appreciated. > > Peter. > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group FieldTrip Development Team P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From r.oostenveld at donders.ru.nl Tue Aug 21 21:56:20 2012 From: r.oostenveld at donders.ru.nl (Robert Oostenveld) Date: Tue, 21 Aug 2012 21:56:20 +0200 Subject: [FieldTrip] source reconstruction on emfs In-Reply-To: <810A8E06C75EB447A8CEB73DBFD7BB0E4C03CD65AD@solaris.neuro.uni-luebeck.de> References: <810A8E06C75EB447A8CEB73DBFD7BB0E4C03CD65AD@solaris.neuro.uni-luebeck.de> Message-ID: Dear Nuria For the earliest components from S1, and just to get started and get a feel for your data, a traditional dipole fit using FT_DIPOLEFITTING should work fine. Following the initial localization of S1, you indeed could continue with an LCMV beamformer. There is an example at http://fieldtrip.fcdonders.nl/example/fit_a_dipole_to_the_tactile_erf_after_mechanical_stimulation The example is on CTF data with Braille-cell stimulation (i.e. mechanical instead of electrical),. Since the example data is available from the ftp you can try the example out before translating it to your own data. Furthermore I should note that the example is rather old and probably has not been run for quite some time, so there might be some pieces of the example script on the wiki that need updating to reflect the current state of the fieldtrip code. I suggest you also have a look here http://fieldtrip.fcdonders.nl/tutorial/headmodel_meg, which is a new tutorial that Lilla Magyari (CC) made available last week. best regards Robert On 21 Aug 2012, at 12:50, Nuria Doñamayor Alonso wrote: > Dear fieldtrippers, > I am currently analyzing a study, measured with a Neuromag 306 system, in which we stimulated the median nerve. I had never done source reconstruction with fieldtrip, so I followed the tutorial using MNE for EMF source analysis (which worked fine), only to find out that I could neither average individual sources nor perform stats, which is exactly what I need to do... > So the question is: What would you guys recommend? Should I just try using the LCMV beamfomer? However, that might not be the best solution according to this thread http://mailman.science.ru.nl/pipermail/fieldtrip/2011-September/004321.html... Is there any way I could efficiently export my preprocessed/averaged data to some other software (MNE/SPM/Brainstorm...) and compute my source reconstruction and stats there? What do you guys usually do? > Thanks a lot for your help, > Nuria > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From Hisako.Fujiwara at cchmc.org Tue Aug 21 22:13:04 2012 From: Hisako.Fujiwara at cchmc.org (Fujiwara, Hisako) Date: Tue, 21 Aug 2012 20:13:04 +0000 Subject: [FieldTrip] beamformer application to spontaneous signal source construction Message-ID: Dear all, I have a spontaneous recording of a patient who has epilepsy. I want to use beamformer to reconstruct and display an increase power at certain frequency and location in the brain. Your example only describe how to use beamfomer for ERP. Could you kindly advise how to apply your fieldtrip beamformer method for my needed application. I would like to thank your all kind help and advice in advance. Sincerely, Hisako -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Wed Aug 22 19:59:51 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Wed, 22 Aug 2012 19:59:51 +0200 Subject: [FieldTrip] FT_REJECTARTIFACT Message-ID: Dear all, I have one problem concerning ft_rejectartifact function. I did everything as you explained in automatic artifact rejection tutorial. I mark segments first and that is fine, but the ft_rejectartifact doesn't remove marked bad segments form my data. Please help! Here is my code: % automatic clear all s=[87 209 195 225 60]; for g=1:length(s) input_no=s(g) clear sub sub=['Subject' num2str(input_no)]; eval(sub) %calling subject's script cfg=[]; cfg.trialdef.eventtype = 'UPPT001'; cfg.trialdef.eventvalue = 1; cfg.trialdef.prestim = 0.1; cfg.trialdef.poststim = 0.5; cfg.trialfun = 'trialfun_general'; cfg.channel = {'MEG'}; cfg.continuous = 'yes'; cfg.dataset = [subjectdata.wr11.datadir]; [data1]=ft_definetrial(cfg) trial=data1.trl; % muscle cfg=[]; cfg.trl =trial; cfg.datafile = [subjectdata.wr11.datafile]; cfg.headerfile =[subjectdata.wr11.headerfile]; cfg.continuous = 'yes'; % channel selection, cutoff and padding cfg.artfctdef.zvalue.channel = 'MRT*'; cfg.artfctdef.zvalue.cutoff = 15; cfg.artfctdef.zvalue.trlpadding = 0.5; cfg.artfctdef.zvalue.fltpadding = 0.5; cfg.artfctdef.zvalue.artpadding = 0.1; % algorithmic parameters cfg.artfctdef.zvalue.bpfilter = 'yes'; cfg.artfctdef.zvalue.bpfreq = [110 140]; cfg.artfctdef.zvalue.bpfiltord = 9; cfg.artfctdef.zvalue.bpfilttype = 'but'; cfg.artfctdef.zvalue.hilbert = 'yes'; cfg.artfctdef.zvalue.boxcar = 0.2; %make the process interactive cfg.artfctdef.zvalue.interactive = 'yes'; [cfg, artifact_muscle] = ft_artifact_zvalue(cfg) % jump cfg=[]; cfg.trl = trial; cfg.datafile = [subjectdata.wr11.datafile]; cfg.headerfile =[subjectdata.wr11.headerfile]; cfg.continuous = 'yes'; % channel selection, cutoff and padding cfg.artfctdef.zvalue.channel = 'MEG'; cfg.artfctdef.zvalue.cutoff = 20; cfg.artfctdef.zvalue.trlpadding = 0; cfg.artfctdef.zvalue.artpadding = 0; cfg.artfctdef.zvalue.fltpadding = 0; % algorithmic parameters cfg.artfctdef.zvalue.cumulative = 'yes'; cfg.artfctdef.zvalue.medianfilter = 'yes'; cfg.artfctdef.zvalue.medianfiltord = 9; cfg.artfctdef.zvalue.absdiff = 'yes'; % make the process interactive cfg.artfctdef.zvalue.interactive = 'yes'; [cfg, artifact_jump] = ft_artifact_zvalue(cfg) cfg.artfctdef.reject = 'complete'; cfg.artfctdef.jump.artifact = artifact_jump; cfg.artfctdef.muscle.artifact = artifact_muscle; data_no_artifacts = ft_rejectartifact(data1) end Thank you in advance! NP -------------- next part -------------- An HTML attachment was scrubbed... URL: From eelke.spaak at donders.ru.nl Wed Aug 22 20:17:29 2012 From: eelke.spaak at donders.ru.nl (Eelke Spaak) Date: Wed, 22 Aug 2012 20:17:29 +0200 Subject: [FieldTrip] FT_REJECTARTIFACT In-Reply-To: References: Message-ID: Dear Nenad, My guess is you are missing the first argument in your call to ft_rejectartifact; i.e. instead of data_no_artifacts = ft_rejectartifact(data1) you should use data_no_artifacts = ft_rejectartifact(cfg, data1) Does that work? Best, Eelke On 22 August 2012 19:59, Nenad Polomac wrote: > Dear all, > > I have one problem concerning ft_rejectartifact function. > I did everything as you explained in automatic artifact rejection tutorial. > I mark segments first and that is fine, but the ft_rejectartifact doesn't > remove marked bad segments form my data. > Please help! > > Here is my code: > > % automatic > clear all > > s=[87 209 195 225 60]; > > > for g=1:length(s) > input_no=s(g) > > clear sub > > sub=['Subject' num2str(input_no)]; > eval(sub) %calling subject's script > > > cfg=[]; > cfg.trialdef.eventtype = 'UPPT001'; > cfg.trialdef.eventvalue = 1; > cfg.trialdef.prestim = 0.1; > cfg.trialdef.poststim = 0.5; > cfg.trialfun = 'trialfun_general'; > cfg.channel = {'MEG'}; > cfg.continuous = 'yes'; > cfg.dataset = [subjectdata.wr11.datadir]; > > [data1]=ft_definetrial(cfg) > trial=data1.trl; > > > % muscle > cfg=[]; > cfg.trl =trial; > cfg.datafile = [subjectdata.wr11.datafile]; > cfg.headerfile =[subjectdata.wr11.headerfile]; > cfg.continuous = 'yes'; > > % channel selection, cutoff and padding > cfg.artfctdef.zvalue.channel = 'MRT*'; > cfg.artfctdef.zvalue.cutoff = 15; > cfg.artfctdef.zvalue.trlpadding = 0.5; > cfg.artfctdef.zvalue.fltpadding = 0.5; > cfg.artfctdef.zvalue.artpadding = 0.1; > > % algorithmic parameters > cfg.artfctdef.zvalue.bpfilter = 'yes'; > cfg.artfctdef.zvalue.bpfreq = [110 140]; > cfg.artfctdef.zvalue.bpfiltord = 9; > cfg.artfctdef.zvalue.bpfilttype = 'but'; > cfg.artfctdef.zvalue.hilbert = 'yes'; > cfg.artfctdef.zvalue.boxcar = 0.2; > > %make the process interactive > cfg.artfctdef.zvalue.interactive = 'yes'; > > [cfg, artifact_muscle] = ft_artifact_zvalue(cfg) > > > > % jump > cfg=[]; > cfg.trl = trial; > cfg.datafile = [subjectdata.wr11.datafile]; > cfg.headerfile =[subjectdata.wr11.headerfile]; > cfg.continuous = 'yes'; > > % channel selection, cutoff and padding > cfg.artfctdef.zvalue.channel = 'MEG'; > cfg.artfctdef.zvalue.cutoff = 20; > cfg.artfctdef.zvalue.trlpadding = 0; > cfg.artfctdef.zvalue.artpadding = 0; > cfg.artfctdef.zvalue.fltpadding = 0; > > % algorithmic parameters > cfg.artfctdef.zvalue.cumulative = 'yes'; > cfg.artfctdef.zvalue.medianfilter = 'yes'; > cfg.artfctdef.zvalue.medianfiltord = 9; > cfg.artfctdef.zvalue.absdiff = 'yes'; > > % make the process interactive > cfg.artfctdef.zvalue.interactive = 'yes'; > > [cfg, artifact_jump] = ft_artifact_zvalue(cfg) > > > > > cfg.artfctdef.reject = 'complete'; > cfg.artfctdef.jump.artifact = artifact_jump; > cfg.artfctdef.muscle.artifact = artifact_muscle; > data_no_artifacts = ft_rejectartifact(data1) > end > > Thank you in advance! > NP > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From polomacnenad at gmail.com Thu Aug 23 10:50:20 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Thu, 23 Aug 2012 10:50:20 +0200 Subject: [FieldTrip] FT_REJECTARTIFACT Message-ID: Dear Eelke, Thank you for your suggestion. But unfortunately this doesn't work. If I do your way I get: Error using ft_rejectartifact (line 233) no trials were selected, cannot perform artifact detection/rejection. So I am still not sure what might be the problem... Regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From stan.vanpelt at fcdonders.ru.nl Thu Aug 23 11:25:21 2012 From: stan.vanpelt at fcdonders.ru.nl (Stan van Pelt) Date: Thu, 23 Aug 2012 11:25:21 +0200 (CEST) Subject: [FieldTrip] FT_REJECTARTIFACT In-Reply-To: Message-ID: <1052918099.53046.1345713921693.JavaMail.root@sculptor.zimbra.ru.nl> Dear Nenad, It seems that no trials are defined when you run ft_definetrial. Can you check if there are actually is trial data present in your data1 structure? If there is no data, you should check if your configuration settings are correct in that processing step (e.g., trigger channel, event values). Best, Stan ----- Oorspronkelijk bericht ----- > Van: "Nenad Polomac" > Aan: fieldtrip at science.ru.nl > Verzonden: Donderdag 23 augustus 2012 10:50:20 > Onderwerp: Re: [FieldTrip] FT_REJECTARTIFACT > Dear Eelke, > Thank you for your suggestion. But unfortunately this doesn't work. If > I do your way I get: Error using ft_rejectartifact (line 233) > no trials were selected, cannot perform artifact detection/rejection. > So I am still not sure what might be the problem... > Regards! > Nenad > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Stan van Pelt Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen Kapittelweg 29, 6525 EN Nijmegen, Netherlands E-mail: stan.vanpelt at donders.ru.nl Website: www.ru.nl/donders/ Phone: (+31) (0)24 36 10981 Fax: (+31) (0)24 36 10989 -------------- next part -------------- An HTML attachment was scrubbed... URL: From poortjasper at gmail.com Thu Aug 23 12:34:48 2012 From: poortjasper at gmail.com (Jasper Poort) Date: Thu, 23 Aug 2012 11:34:48 +0100 Subject: [FieldTrip] Fwd: ft_freqdescriptives and cfg.trial In-Reply-To: References: Message-ID: Dear Fieldtrip users, I have a problem with using a trial selection to do freqdescriptives on a subset of trials: I first do ft_freqanalysis with these setttings, which results in freq: cfg.output = 'fourier'; cfg.method = 'mtmconvol'; cfg.keeptrials = 'yes'; freq = label: {1x64 cell} dimord: 'rpttap_chan_freq_time' freq: [30 35 40 45 50 55 60 65 70 75 80 85 90 95 100] time: [1x36 double] fourierspctrm: [4-D double] cumtapcnt: [857x15 double] cfg: [1x1 struct] size(freq.fourierspctrm) ans = 2571 64 15 36 then i try running ft_freqdescriptives with these settings tmpcfg = []; tmpcfg.jackknife = 'no'; tmpcfg.keeptrials = 'no'; tmpcfg.channel = freq.label([chnix1,chnix2]) tmpcfg.trials = [1:10:100]; tmp = ft_freqdescriptives(tmpcfg,freq); Error using + Array dimensions must match for binary array op. Error in ft_checkdata>fixcsd (line 740) powspctrm = powspctrm + abs(data.fourierspctrm(p:ntap:end,:,:,:,:)).^2; Error in ft_checkdata (line 646) data = fixcsd(data, cmbrepresentation, channelcmb); Error in ft_freqdescriptives (line 131) freq = ft_checkdata(freq, 'cmbrepresentation', 'sparsewithpow', 'channelcmb', {}); the error seems to happen because the cumtapcnt seems to be updated after calling ft_selectdata in ft_freqdescriptives such that it no longer contains a row for every trial which causes nrpt = size(data.cumtapcnt,1); in ft_checkdata to return 1 instead of the number of trials. label: {2x1 cell} dimord: 'rpttap_chan_freq_time' freq: [30 35 40 45 50 55 60 65 70 75 80 85 90 95 100] time: [1x36 double] fourierspctrm: [4-D double] cumtapcnt: [3 3 3 3 3 3 3 3 3 3] cfg: [1x1 struct] I was wondering if this is a bug or am I somehow specifying the input incorrectly? Thanks! best, Jasper -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 23 12:55:02 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 23 Aug 2012 12:55:02 +0200 Subject: [FieldTrip] Fwd: ft_freqdescriptives and cfg.trial In-Reply-To: References: Message-ID: Hi Jasper, I think this is indeed a bug in fieldtrip. The reason being that the cumtapcnt field for timefrequency data is not well-defined/well-designed/well-handled (the latter by ft_selectdata, which is doing the trial selection). Probably you can work around this by updating the cumtapcnt field prior to calling ft_freqdescriptives. I suspect that it will work if you define it to be a vector (nx1) where n is the number of trials, and each entry has the value of the number of tapers applied (per trial). Would it be possible for you to create a bug on our bugzilla site (bugzilla.fcdonders.nl) with a description of the problem, a little test script and some (small) data for reproduction? The team will then take it from there. Best, Jan-Mathijs On Aug 23, 2012, at 12:34 PM, Jasper Poort wrote: > Dear Fieldtrip users, > > > > I have a problem with using a trial selection to do freqdescriptives on a subset of trials: > > > > I first do ft_freqanalysis with these setttings, which results in freq: > > > > cfg.output = 'fourier'; > > cfg.method = 'mtmconvol'; > > cfg.keeptrials = 'yes'; > > > freq = > > > label: {1x64 cell} > > dimord: 'rpttap_chan_freq_time' > > freq: [30 35 40 45 50 55 60 65 70 75 80 85 90 95 100] > > time: [1x36 double] > > fourierspctrm: [4-D double] > > cumtapcnt: [857x15 double] > > cfg: [1x1 struct] > > > size(freq.fourierspctrm) > > ans = > > 2571 64 15 36 > > > then i try running ft_freqdescriptives with these settings > > tmpcfg = []; > > tmpcfg.jackknife = 'no'; > > tmpcfg.keeptrials = 'no'; > > tmpcfg.channel = freq.label([chnix1,chnix2]) > > tmpcfg.trials = [1:10:100]; > > tmp = ft_freqdescriptives(tmpcfg,freq); > > > Error using + > > Array dimensions must match for binary array op. > > > Error in ft_checkdata>fixcsd (line 740) > > powspctrm = powspctrm + abs(data.fourierspctrm(p:ntap:end,:,:,:,:)).^2; > > > Error in ft_checkdata (line 646) > > data = fixcsd(data, cmbrepresentation, channelcmb); > > > Error in ft_freqdescriptives (line 131) > > freq = ft_checkdata(freq, 'cmbrepresentation', 'sparsewithpow', 'channelcmb', {}); > > > > the error seems to happen because the cumtapcnt seems to be updated after calling ft_selectdata in ft_freqdescriptives such that it no longer contains a row for every trial which causes nrpt = size(data.cumtapcnt,1); in ft_checkdata to return 1 instead of the number of trials. > > > label: {2x1 cell} > dimord: 'rpttap_chan_freq_time' > freq: [30 35 40 45 50 55 60 65 70 75 80 85 90 95 100] > time: [1x36 double] > fourierspctrm: [4-D double] > cumtapcnt: [3 3 3 3 3 3 3 3 3 3] > cfg: [1x1 struct] > > > > I was wondering if this is a bug or am I somehow specifying the input incorrectly? > > > > Thanks! best, Jasper > > > > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jm.horschig at donders.ru.nl Thu Aug 23 13:41:00 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Thu, 23 Aug 2012 13:41:00 +0200 Subject: [FieldTrip] FT_REJECTARTIFACT In-Reply-To: References: Message-ID: <503616CC.7010303@donders.ru.nl> Dear Nenad, it is a bit confusing that you call the variable returned from ft_definetrial 'data1', because actually it is only a cfg-structure without data. No matter how you call that variable though, it should work if you use this: / data1.artfctdef.reject = 'complete'; / / data1.artfctdef.jump.artifact = artifact_jump;/ / data1.artfctdef.muscle.artifact = artifact_muscle;/ / data_no_artifacts = ft_rejectartifact(data1)/ Note that you need to call ft_preprocessing in order to obtain data, in your case: /data = ft_preprocessing(data_no_artifacts);/ or /data = ft_preprocessing(data1);/ ft_definetrial is a function which just returns at what samples your trials are supposed to start. ft_preprocessing is then using this information to extract the trials out of your datafile. You can artifact rejection either before you segmented the data into trials, as done here, or afterwards, i.e. after having called ft_preprocessing. Best, Jörn On 8/23/2012 10:50 AM, Nenad Polomac wrote: > Dear Eelke, > > Thank you for your suggestion. But unfortunately this doesn't work. If > I do your way I get: Error using ft_rejectartifact (line 233) > no trials were selected, cannot perform artifact detection/rejection. > > So I am still not sure what might be the problem... > > Regards! > > Nenad > > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group FieldTrip Development Team P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Thu Aug 23 14:46:37 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Thu, 23 Aug 2012 14:46:37 +0200 Subject: [FieldTrip] FT_REJECTARTIFACT In-Reply-To: References: Message-ID: Dear Jörn, Thank you for your answer! That was the problem. I understand all now. Regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From s.rombetto at cib.na.cnr.it Thu Aug 23 14:57:51 2012 From: s.rombetto at cib.na.cnr.it (s.rombetto at cib.na.cnr.it) Date: Thu, 23 Aug 2012 14:57:51 +0200 Subject: [FieldTrip] EDF data Message-ID: <20120823145751.avcc2gkcw0ws0wow@arco.cib.na.cnr.it> Dear fieldtrip users, I'm trying to import some EEG data in EDF format. I am able to read the data structure using hdr= ft_read_header(rawdataname); and also to read the data by data = ft_read_data(rawdataname);, but I receive an error message when I try to perform the ft_preprocessing command (Error in ==> ft_preprocessing at 511). Do you have any hint or sample code? Thank you in advance! Sara ------------------------- Dott.ssa Sara Rombetto Istituto di Cibernetica "E. Caianiello" Via Campi Flegrei, 34 80078 Pozzuoli (NA) Italy tel +390818675361 fax +390818675128 -------------------------- "I disapprove of what you say, but I will defend to the death your right to say it." [Evelyn Beatrice Hall, The Friends Of Voltaire] ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From nuria.donamayor at neuro.uni-luebeck.de Thu Aug 23 16:03:04 2012 From: nuria.donamayor at neuro.uni-luebeck.de (=?iso-8859-1?Q?Nuria_Do=F1amayor_Alonso?=) Date: Thu, 23 Aug 2012 16:03:04 +0200 Subject: [FieldTrip] source reconstruction on emfs In-Reply-To: References: <810A8E06C75EB447A8CEB73DBFD7BB0E4C03CD65AD@solaris.neuro.uni-luebeck.de>, Message-ID: <810A8E06C75EB447A8CEB73DBFD7BB0E4C03CD65C1@solaris.neuro.uni-luebeck.de> Dear Robert, I did what you suggested and everything worked perfectly, thanks. Now I have a slight problem with the following. I want to delimit a ROI to do stats on my LCMV results and, as far as I understand, I should use ft_sourceinterpolate, ft_volumenormalise and then ft_sourcegrandaverage. The problem is, when I try to check out the result with ft_sourceplot, it just plots the first slice 20 times. This does not happen with the non-normalised grandaverages (and the individual averages in both cases). Any suggestions? Here's the code I'm using: # on the individual data cfg = []; cdf.downsample = 2; cfg.parameter = 'avg.nai'; sourceint = ft_sourceinterpolate(cfg, source, mri); cfg = []; cfg.coordsys = 'neuromag'; cfg.nonlinear = 'no'; sourcenorm = ft_volumenormalise(cfg, sourceint); # grandaverage cfg = []; cfg.keepindividual = 'yes'; cfg.parameter = 'nai'; gsource = (cfg, sourceint1, sourceint2...); # plot gsource cfg = []; cfg.funparameter = 'avg.nai'; cfg.maskparameter = cfg.funparameter; ft_sourceplot(cfg, gsource); Thanks again! Nuria ________________________________________ Von: fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl] im Auftrag von Robert Oostenveld [r.oostenveld at donders.ru.nl] Gesendet: Dienstag, 21. August 2012 21:56 An: FieldTrip discussion list Betreff: Re: [FieldTrip] source reconstruction on emfs Dear Nuria For the earliest components from S1, and just to get started and get a feel for your data, a traditional dipole fit using FT_DIPOLEFITTING should work fine. Following the initial localization of S1, you indeed could continue with an LCMV beamformer. There is an example at http://fieldtrip.fcdonders.nl/example/fit_a_dipole_to_the_tactile_erf_after_mechanical_stimulation The example is on CTF data with Braille-cell stimulation (i.e. mechanical instead of electrical),. Since the example data is available from the ftp you can try the example out before translating it to your own data. Furthermore I should note that the example is rather old and probably has not been run for quite some time, so there might be some pieces of the example script on the wiki that need updating to reflect the current state of the fieldtrip code. I suggest you also have a look here http://fieldtrip.fcdonders.nl/tutorial/headmodel_meg, which is a new tutorial that Lilla Magyari (CC) made available last week. best regards Robert On 21 Aug 2012, at 12:50, Nuria Doñamayor Alonso wrote: > Dear fieldtrippers, > I am currently analyzing a study, measured with a Neuromag 306 system, in which we stimulated the median nerve. I had never done source reconstruction with fieldtrip, so I followed the tutorial using MNE for EMF source analysis (which worked fine), only to find out that I could neither average individual sources nor perform stats, which is exactly what I need to do... > So the question is: What would you guys recommend? Should I just try using the LCMV beamfomer? However, that might not be the best solution according to this thread http://mailman.science.ru.nl/pipermail/fieldtrip/2011-September/004321.html... Is there any way I could efficiently export my preprocessed/averaged data to some other software (MNE/SPM/Brainstorm...) and compute my source reconstruction and stats there? What do you guys usually do? > Thanks a lot for your help, > Nuria > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip _______________________________________________ fieldtrip mailing list fieldtrip at donders.ru.nl http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From kamx at ymail.com Fri Aug 24 06:56:53 2012 From: kamx at ymail.com (Kam X) Date: Thu, 23 Aug 2012 21:56:53 -0700 (PDT) Subject: [FieldTrip] Problem setting up fieldtrip Message-ID: <1345784213.64205.YahooMailClassic@web162305.mail.bf1.yahoo.com> Hi, I hope this is the proper type of questions for this list, if not I apologize.I've just set up fieldtrip by unzipping it, adding the path to the folder to matlab, and running ft_defaults.  Most things seem to work, but if I try to use the command ft_realtime_signalproxy I get: Undefined function 'ft_realtime_signalproxy' for input arguments of type 'struct'. Looking in the fieldtrip folder, I see the .m files, but ft_realtime_signalproxy is not among them.  If I run realtime_signalproxy(cfg), I get: Warning: REALTIME_SIGNALPROXY is only a compatibility wrapper, which will soon be removed.Please instead call FT_REALTIME_SIGNALPROXY. > In realtime_signalproxy at 16 Undefined function 'ft_realtime_signalproxy' for input arguments of type 'struct'. Error in realtime_signalproxy (line 18)[varargout{1:nargout}] = funhandle(varargin{:}); Have I missed a step in setting up fieldtrip, or do I need to download the realtime functions from somewhere? Thanks,Kameron -------------- next part -------------- An HTML attachment was scrubbed... URL: From jm.horschig at donders.ru.nl Fri Aug 24 09:51:06 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Fri, 24 Aug 2012 09:51:06 +0200 Subject: [FieldTrip] Problem setting up fieldtrip In-Reply-To: <1345784213.64205.YahooMailClassic@web162305.mail.bf1.yahoo.com> References: <1345784213.64205.YahooMailClassic@web162305.mail.bf1.yahoo.com> Message-ID: <5037326A.8030209@donders.ru.nl> Dear Kameron, Thanks for reporting this. FieldTrip is modularly organized, with 'realtime' being a separate module/toolbox. Usually, modules should be added automatically when calling ft_defaults. The ft_hastoolbox function should take care of this. However, the realtime folder has been restructured recently so that it has subfolders now, which are (unfortunately) not added by default. It is a(tiny little) bit more tedious now to add this to your path manually: /ft_hastoolbox('realtime/example', 1)/ The same holds for any other subfolders of the realtime module. I'm gonna check with Robert (probably after the BioMag conference) how to resolve this best (if at all). Best, Jörn On 8/24/2012 6:56 AM, Kam X wrote: > Hi, > > I hope this is the proper type of questions for this list, if not I > apologize. > I've just set up fieldtrip by unzipping it, adding the path to the > folder to matlab, and running ft_defaults. Most things seem to work, > but if I try to use the command ft_realtime_signalproxy I get: > > Undefined function 'ft_realtime_signalproxy' for input arguments of > type 'struct'. > > Looking in the fieldtrip folder, I see the .m files, but > ft_realtime_signalproxy is not among them. If I run > realtime_signalproxy(cfg), I get: > > Warning: REALTIME_SIGNALPROXY is only a compatibility wrapper, which > will soon be removed. > Please instead call FT_REALTIME_SIGNALPROXY. > > In realtime_signalproxy at 16 > Undefined function 'ft_realtime_signalproxy' for input arguments of > type 'struct'. > > Error in realtime_signalproxy (line 18) > [varargout{1:nargout}] = funhandle(varargin{:}); > > Have I missed a step in setting up fieldtrip, or do I need to download > the realtime functions from somewhere? > > Thanks, > Kameron > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group FieldTrip Development Team P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From bobrobbrown at googlemail.com Fri Aug 24 10:38:58 2012 From: bobrobbrown at googlemail.com (Robert Brown) Date: Fri, 24 Aug 2012 11:38:58 +0300 Subject: [FieldTrip] PLV on sources Message-ID: Dear Fieldtrippers, I see (from the ft_connectivityanalysis code) that one cannot run phase locking (PLV) analysis (a la Lachaux et al) on source data. My brain still insisted on trying to do so but unfortunately I am not sure if it is clever enough. Hope you can help! It seems that a way to go would be to do LCMV beamforming on the data, get the virtual electrodes, then run freqanalysis with fourier output on these virtual electrodes and then perform PLV on those data. Before I dig myself very deep into this, maybe some experts here could guide me: Would this approach work? Any possible caveats? Are there maybe other, better and more straightforward ways for achieving the PLV on source level? Any comments would be appreciated. Thank you very much for your time! Kind regards, Bob -------------- next part -------------- An HTML attachment was scrubbed... URL: From s.bogels at psy.gla.ac.uk Fri Aug 24 13:25:48 2012 From: s.bogels at psy.gla.ac.uk (Sara =?iso-8859-1?b?QvZnZWxz?=) Date: Fri, 24 Aug 2012 12:25:48 +0100 Subject: [FieldTrip] How to specify two dipoles in cfg.supdip in ft_sourceanalysis? Message-ID: <20120824122548.739016jhmbe93sn0@horde.psy.gla.ac.uk> Hi all, I am using ft_sourceanalysis with specifications of cfg.refdip and cfg.supdip. Until now, I always inserted the same voxel coordinates in supdip as refdip to suppress self-coherence. However, the help suggests that you can also suppress more than one dipole, but it does not specify how to input this in cfg.supdip. I tried a 2 by 3 matrix and a cell structure with two entries (1 x 3 matrices) but that gave (apparently) the same result as suppressing only refdip. As a related, more general question, what about the orientation of the (e.g. reference) dipole? How is that actually determined in this case (since you only have to insert the position of the dipole)? Thank you, Sara ---------------------------------------------------------------- This message was sent using the Web mail system for The University of Glasgow School of Psychology ------------------------------------------------------------------ From narayan.ps at tut.fi Fri Aug 24 14:35:40 2012 From: narayan.ps at tut.fi (Narayan Puthanmadam Subramaniyam) Date: Fri, 24 Aug 2012 12:35:40 +0000 Subject: [FieldTrip] PLV on sources Message-ID: <201208241235.q7OCZhaZ002930@brievenbus.science.ru.nl> hi i was thinking of doing the same thing exactly. Computing plv on the eeg inverse solution. technically it should be possible going by the general level concept of phase locking. but i am not sure so let the experts here speak. Thanks&Regards, NPS Sent from my Nokia phone -----Original Message----- From: Robert Brown Sent: 24:08:2012, 11:38 To: fieldtrip at science.ru.nl Subject: [FieldTrip] PLV on sources Dear Fieldtrippers, I see (from the ft_connectivityanalysis code) that one cannot run phase locking (PLV) analysis (a la Lachaux et al) on source data. My brain still insisted on trying to do so but unfortunately I am not sure if it is clever enough. Hope you can help! It seems that a way to go would be to do LCMV beamforming on the data, get the virtual electrodes, then run freqanalysis with fourier output on these virtual electrodes and then perform PLV on those data. Before I dig myself very deep into this, maybe some experts here could guide me: Would this approach work? Any possible caveats? Are there maybe other, better and more straightforward ways for achieving the PLV on source level? Any comments would be appreciated. Thank you very much for your time! Kind regards, Bob _______________________________________________ fieldtrip mailing list fieldtrip at donders.ru.nl http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From maximilien.chaumon at gmail.com Fri Aug 24 17:13:22 2012 From: maximilien.chaumon at gmail.com (Maximilien Chaumon) Date: Fri, 24 Aug 2012 17:13:22 +0200 Subject: [FieldTrip] [Eeglablist] nested hypothesis testing to decide whether to use one or two dipoles to fit a component In-Reply-To: References: Message-ID: Hi everyone, Just to follow-up on this if anyone wanted to take that approach. I did use the following procedure to decide whether to fit one or two dipoles to a given component. 1) fit one dipole, compute SSE for each component map as the RV * sum of squared inverse weights (sse(1) = EEG.dipfit.model(idip).rv .* sum(EEG.icawinv(:,idip).^2)) 2) fit two dipoles, compute SSE for each component map: sse(2) 3) assume p1 = 6 parameters estimated for 1 dipole, p2 = 9 parameters estimated for 2 dipoles (only three momentum values more, since symmetry imposes position for the two dipoles), n = EEG.nbchan 4) F statistic : F = (sse(1)-sse(2))/(p2-p1)/(sse(2)/(n-p2)) 5) if F > finv(.95,p2-p1,n-p2) decide to go for 2 dipoles. 6) that yielded some solutions where 2 dipoles were selected but were in fact at the same position (x == 0) with opposite momenta. Some others had really just one dipole with large amplitude and the other much smaller. So I discarded solutions where distance between the two dipoles was less than 5mm, or where the momentum angles were opposed, or when the amplitude ratio between the two dipoles was bigger than 5. The final solution is only moderately satisfying. Looking at the maps, the 2 dipole solution is still too often favored over the one dipole solution. As Scott previously mentioned, it could be that the second dipole accounts for error in the head model, and that's all. The F difference in SSE is probably not a good measure of the two-dipolarity of a component. One would probably better just screen all components by eye and spot those that look dipolar using thumbometric visual assessment. Best, Max PS: the script I used: dofit = 1; dorework = 0; doplot = 1; rootdir = '/DATA/'; cd(rootdir) studyfile = fullfile(rootdir,'AP2.study'); if dofit || dorework [STUDY ALLEEG] = pop_loadstudy(studyfile); for i_set = 1:numel(STUDY.datasetinfo) EEG = pop_loadset('filename',ALLEEG(i_set).filename,'filepath',ALLEEG(i_set).filepath);%pop_loadset(GRAB(i_set).name); if isfield(EEG,'peakalphafrequency') EEG.peakalphafreq = EEG.peakalphafrequency; EEG = rmfield(EEG,'peakalphafrequency'); end if ~dorework EEG = pop_dipfit_settings( EEG, 'hdmfile',[ fileparts(which('eeglab')) '/plugins/dipfit2.2/standard_BEM/standard_vol.mat'],'coordformat','MNI','mrifile',[ fileparts(which('eeglab')) '/plugins/dipfit2.2/standard_BEM/standard_mri.mat'],'chanfile',[ fileparts(which('eeglab')) '/plugins/dipfit2.2/standard_BEM/elec/standard_1005.elc'],'coord_transform',[0.67468 -15.6307 2.426 0.081417 0.0024349 -1.5728 1.1744 1.0622 1.1487] ,'chansel',[1:EEG.nbchan] ); EEGonedipole = pop_multifit(EEG, [1:size(EEG.icawinv,2)] ,'threshold',100,'rmout','on','dipoles',1); EEGtwodipoles = pop_multifit(EEG, [1:size(EEG.icawinv,2)] ,'threshold',100,'rmout','on','dipoles',2); else EEGonedipole = EEG; EEGtwodipoles = EEG; EEGonedipole.dipfit.model = EEG.dipfit.onedipole; EEGtwodipoles.dipfit.model = EEG.dipfit.twodipoles; end % H0: one dipole. EEG.dipfit = EEGonedipole.dipfit; % still we store them both. doesn't hurt. EEG.dipfit.onedipole = EEGonedipole.dipfit.model; EEG.dipfit.twodipoles = EEGtwodipoles.dipfit.model; % now for every component, if the rv is much better for the % 2dipoles than for the 1dipole, we'll use the 2dipoles. for i_dip = 1:numel(EEGonedipole.dipfit.model) sse(1) = EEGonedipole.dipfit.model(i_dip).rv .* sum(EEGonedipole.icawinv(:,i_dip).^2); sse(2) = EEGtwodipoles.dipfit.model(i_dip).rv .* sum(EEGtwodipoles.icawinv(:,i_dip).^2); p1 = 6;% six params for 1dipole model p2 = 9;% 9 for 2 dipoles model (assume only symmetry is fixed). n = EEG.nbchan; % N is number of electrodes % F statistic F = (sse(1)-sse(2))/(p2-p1)/(sse(2)/(n-p2)); EEG.dipfit.Fone2two(i_dip) = F; % F decision Fthresh = finv(.999,p2-p1,n-p2); if F > Fthresh % reject H0, take 2 dipoles model % only if x ~= 0 and the two momenta are not just opposite. mom = EEGtwodipoles.dipfit.model(i_dip).momxyz; pos = EEGtwodipoles.dipfit.model(i_dip).posxyz; [t p r] = cart2sph(mom(:,1),mom(:,2),mom(:,3)); t(2) = t(2)+pi;% d = sqrt(sum(diff(pos).^2)); % if the two dipoles are too close and have exactly % opposite angles cond = d < 5 && (diff(abs(t)) < 0.01 || diff(abs(p)) < 0.01); % or if they are exactly on the sagittal plane cond = cond | abs(EEGtwodipoles.dipfit.model(i_dip).posxyz(1)) < 5; % or if the norm of one is much bigger than the other ratio = norm(mom(1,:)) ./ norm(mom(2,:)); ratiothresh = 5; cond = cond | (ratio > ratiothresh || ratio < 1/ratiothresh); % we will still keep the one dipole fit if ~cond% so if not, we take the 2 dipole fit. EEG.dipfit.model(i_dip) = EEGtwodipoles.dipfit.model(i_dip); end end end % reject poor overall fits. EEG.dipfit.model = dipfit_reject(EEG.dipfit.model, 20/100); % reject NaN rv for i_m = 1:numel(EEG.dipfit.model) if isnan(EEG.dipfit.model(i_m).rv) EEG.dipfit.model(i_m).posxyz = []; EEG.dipfit.model(i_m).momxyz = []; EEG.dipfit.model(i_m).rv = 1; end end % removing dipoles outside the head % --------------------------------- rmdip = []; for index = 1:numel(EEGonedipole.dipfit.model) if ~isempty(EEG.dipfit.model(index).posxyz) if any(sqrt(sum(EEG.dipfit.model(index).posxyz.^2,2)) > 85) rmdip = [ rmdip index]; EEG.dipfit.model(index).posxyz = []; EEG.dipfit.model(index).momxyz = []; EEG.dipfit.model(index).rv = 1; end; end; end; if ~isempty(rmdip) fprintf('%d out of cortex dipoles removed (usually artifacts)\n', length(rmdip)); end; pop_saveset(EEG,'savemode','resave'); end end %% if doplot % [STUDY ALLEEG] = pop_loadstudy('AlphaPsyc2.study'); num = [];% number of dipoles that are fit for each component idx = []; topos = {[] [] []}; Fdiff = {[] [] []}; idx = {[] [] []}; dips = {}; for i_e = 1:numel(ALLEEG) for i_comp = 1:numel(ALLEEG(i_e).dipfit.model) num(end+1) = size(ALLEEG(i_e).dipfit.model(i_comp).posxyz,1); topos{num(end)+1}(end+1,:) = ALLEEG(i_e).icawinv(:,i_comp)'; dips{num(end)+1}(size(topos{num(end)+1},1)) = ALLEEG(i_e).dipfit.model(i_comp); Fdiff{num(end)+1}(end+1) = ALLEEG(i_e).dipfit.Fone2two(i_comp); idx{num(end)+1}(end+1,:) = [i_e i_comp]; end end for i_plot = 2:3 figure(6548+i_plot);paste_figpos manytopos ntopos = size(topos{i_plot},1); n = ceil(sqrt(ntopos)); m = n; for i_topo = 1:ntopos subplot(n,m,i_topo) cla topoplot(topos{i_plot}(i_topo,:),ALLEEG(1).chanlocs,'conv','on','electrodes','off'); title({num2str(idx{i_plot}(i_topo,:)) num2str(Fdiff{i_plot}(i_topo))}); drawnow end end end 2012/8/13 Maximilien Chaumon > Hello eeglab & Fieldtrip, > > I'm trying to find out if it would be possible to use a nested hypothesis > testing approach to decide whether to use a one or two dipole model while > estimating components' dipole locations. > The rationale I would like to follow is this: with two dipoles, we will > always obtain a better fit than with one dipole, but the decrease in sum of > squared errors (SSE) should follow a F distribution with k (= > Nparameters_2dipoles - Nparameters_1dipole) degrees of freedom. If the > decrease in SSE is greater than what would be expected under this F > distribution, then we decide that 2 dipoles provide a sufficiently better > fit and decide using them. > > I asked this question to eeglablist and Scott pointed out that it is > difficult/impossible(?) to determine if the second dipole fits actual > interesting data or just noise introduced by the imperfect head model. > Christian then said it'd be worth a shot, and I agree, so here I am again > with two questions, or two confirmations: > > 1) *How many parameters are estimated in ft_dipolefitting.m ?* specially > in the case of 2 dipoles. If I count correctly, we estimate 6 parameters > for one dipole, and, depending on whether the orientation has to be the > same in the 2 dipoles, one (amplitude) or three (amplitude and orientation) > more. > 2) *Can I assimilate the relative residual variance to a SSE?* the > function rv.m does this: rv = sum((d1-d2).^2) ./ sum(d1.^2); > So that seems to be a sum of squared errors divided by the variance of the > original data. So if I multiply the rv by the sum squared component map, I > should get it, right? > > Thanks a lot! > Max > > > 2012/8/11 Christian Kothe > >> I can only speak from my armchair here, but it sounds like it should be >> worth a try - even if you don't get the # of parameters exactly right it >> will probably give you at least some level of complexity control in >> whatever the range of validity is. If it works, it may inspire follow-up >> work (e.g., Bayesian model selection or likelihood ratio tests). >> >> The number of parameters for a 2-dipole model seems to be 3 (xyz) + 4 >> (2x the orientation parameters). Not sure about the momentum, though - you >> might look up the place where the actual function minimization is being >> performed in dipfit (fminunc call?) and see whether these are being >> optimized together with the others. >> >> Christian >> >> On Aug 10, 2012, at 3:29 PM, Scott Makeig wrote: >> >> MAX - Unfortunately, in general using two dipoles rather than one will >> ~always improve the fit. Even if the source is a pure single dipole, a >> second dipole can be used to correct for noise or errors in the forward >> head model. This is less always the case for the constrained >> spatially-symmetric dipole pairs allowed by dipfit(). However, we have not >> thought of an optimal way to decide between using one or (occasionally) two >> dipoles to fit e.g. maps of ICA brain sources. The goal would be to decide >> whether the two-dipole version is fitting noise/forward model error vs >> actual bilateral source generation... >> >> Scott Makeig >> >> On Thu, Aug 9, 2012 at 1:54 AM, Maximilien Chaumon < >> maximilien.chaumon at gmail.com> wrote: >> >>> Hello all, >>> >>> When fitting dipoles to components, we are all sooner or later puzzled >>> by the question whether to use one or two symmetrical dipoles. >>> >>> Would it be correct to put the problems in terms of a nested hypothesis >>> testing? >>> >>> We are fitting a scalp map with one or two parameters and get a residual >>> variance after the fit. >>> Could we not use this residual variance as a measure of the SSE and >>> compute a F statistic to decide whether to use the more complex (with two >>> dipoles) or simpler (with one dipole) of two nested models? >>> If yes, then how would we decide on the number of degrees of freedom? >>> How many free parameters do we have in each case? x,y,z,and two >>> orientations per dipole? how does the imposed symmetry affect that number? >>> Could we really map residual variance to SSE? How many "observations" do we >>> have in that case (see formula below)? >>> >>> I found this formula, for F: >>> F = (SSEF-SSER)/ (kF-kR) / ((1-SSEF)/(N-kF-1)) >>> where >>> SSE is sum of squared errors, >>> k is numbers of parameters, >>> N is number of observations (? what in our case?) >>> F and R indices for full and reduced model respectively (in our case two >>> and one dipole). >>> >>> >>> Thanks a lot for any comment! >>> Best, >>> Max >>> >>> dipfit >>> >>> >>> >>> >>> >>> _______________________________________________ >>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html >>> To unsubscribe, send an empty email to >>> eeglablist-unsubscribe at sccn.ucsd.edu >>> For digest mode, send an email with the subject "set digest mime" to >>> eeglablist-request at sccn.ucsd.edu >>> >> >> >> >> -- >> Scott Makeig, Research Scientist and Director, Swartz Center for >> Computational Neuroscience, Institute for Neural Computation; Prof. of >> Neurosciences (Adj.), University of California San Diego, La Jolla CA >> 92093-0559, http://sccn.ucsd.edu/~scott >> >> _______________________________________________ >> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html >> To unsubscribe, send an empty email to >> eeglablist-unsubscribe at sccn.ucsd.edu >> For digest mode, send an email with the subject "set digest mime" to >> eeglablist-request at sccn.ucsd.edu >> >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Fri Aug 24 19:51:15 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Fri, 24 Aug 2012 19:51:15 +0200 Subject: [FieldTrip] ft_databrowser Message-ID: Hi, Does anybody knows why I can't see whole head topography in ft_databrowser cfg = []; cfg.viewmode ='component'; cfg.layout = 'CTF275.lay'; ft_databrowser(cfg, comp); [image: Inline images 1] Kind regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.jpeg Type: image/jpeg Size: 78272 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ica.JPG Type: image/jpeg Size: 78272 bytes Desc: not available URL: From r.vandermeij at donders.ru.nl Sat Aug 25 11:40:08 2012 From: r.vandermeij at donders.ru.nl (Roemer van der Meij) Date: Sat, 25 Aug 2012 11:40:08 +0200 Subject: [FieldTrip] ft_databrowser In-Reply-To: References: Message-ID: Hi Nenad, This was indeed a bug, and was fixed very recently. If you update to the latest version, it should work fine. Do note the speed-upgrade that I implemented ;), it should be much faster now. Also, just to gather some interest, when you now make a data-selection in the browser, you can right-click to get a context menu with several interesting options, do have a look! :) Best, Roemer On Fri, Aug 24, 2012 at 7:51 PM, Nenad Polomac wrote: > Hi, > > Does anybody knows why I can't see whole head topography in ft_databrowser > > > cfg = []; > cfg.viewmode ='component'; > cfg.layout = 'CTF275.lay'; > ft_databrowser(cfg, comp); > > [image: Inline images 1] > > Kind regards! > > Nenad > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- Roemer van der Meij M.Sc. PhD student Donders Institute for Brain, Cognition and Behaviour Centre for Cognition P.O. Box 9104 6500 HE Nijmegen The Netherlands Tel: +31(0)24 3655932 E-mail: r.vandermeij at donders.ru.nl -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.jpeg Type: image/jpeg Size: 78272 bytes Desc: not available URL: From polomacnenad at gmail.com Sat Aug 25 20:03:09 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Sat, 25 Aug 2012 20:03:09 +0200 Subject: [FieldTrip] fieldtrip Digest, Vol 21, Issue 31 In-Reply-To: References: Message-ID: Dear Roemer, Thank you very much for your suggestion. Now works fine and much faster. Keep up the good work! :) All the best! Nenad > Hi Nenad, > > This was indeed a bug, and was fixed very recently. If you update to the > latest version, it should work fine. Do note the speed-upgrade that I > implemented ;), it should be much faster now. > Also, just to gather some interest, when you now make a data-selection in > the browser, you can right-click to get a context menu with several > interesting options, do have a look! :) > > Best, > Roemer > > > > On Fri, Aug 24, 2012 at 7:51 PM, Nenad Polomac >wrote: > > > Hi, > > > > Does anybody knows why I can't see whole head topography in > ft_databrowser > > > > > > cfg = []; > > cfg.viewmode ='component'; > > cfg.layout = 'CTF275.lay'; > > ft_databrowser(cfg, comp); > > > > [image: Inline images 1] > > > > Kind regards! > > > > Nenad > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > > -- > Roemer van der Meij M.Sc. > PhD student > Donders Institute for Brain, Cognition and Behaviour > Centre for Cognition > P.O. Box 9104 > 6500 HE Nijmegen > The Netherlands > Tel: +31(0)24 3655932 > E-mail: r.vandermeij at donders.ru.nl > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: < > http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20120825/351f7b56/attachment.html > > > -------------- next part -------------- > A non-text attachment was scrubbed... > Name: not available > Type: image/jpeg > Size: 78272 bytes > Desc: not available > URL: < > http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20120825/351f7b56/attachment.jpe > > > > ------------------------------ > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > End of fieldtrip Digest, Vol 21, Issue 31 > ***************************************** > -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Sat Aug 25 20:09:30 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Sat, 25 Aug 2012 20:09:30 +0200 Subject: [FieldTrip] ft_databrowser Message-ID: Dear Roemer, Thank you very much for your suggestion. Now works fine and much faster. Keep up the good work! :) I am sorry but I don't know how to put replay massage in the same thread in the Fieldtrip discussion list . Please give me some clue All the best! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From ayelet.landau at gmail.com Sat Aug 25 21:03:46 2012 From: ayelet.landau at gmail.com (Ayelet Landau) Date: Sat, 25 Aug 2012 21:03:46 +0200 Subject: [FieldTrip] ft_databrowser In-Reply-To: References: Message-ID: Hi all, I thought I would report here also another (perhaps related?) strange behavior that I encountered with databrowser. If you look at the attached image you will immediately note that there is an empty portion after the end of my data. The data plotted is indeed all the data (so, no missing data), but this extra white just appears to varying degrees from trial to trial. any idea why this occurs? thanks! Ayelet [image: Inline image 1] On Sat, Aug 25, 2012 at 8:09 PM, Nenad Polomac wrote: > > Dear Roemer, > > Thank you very much for your suggestion. Now works fine and much faster. > Keep up the good work! :) > > I am sorry but I don't know how to put replay massage in the same thread > in the Fieldtrip discussion list . Please give me some clue > > All the best! > > Nenad > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- Ayelet N. Landau Postdoctoral Scientist, Ernst Strüngmann Institute (ESI) for Neuroscience in Cooperation with Max Planck Society Deutschordenstr. 46, D-60528 Frankfurt Mobile: +49 (0)16 22733 110 Fax: +49 (0)69 96769 555 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.png Type: image/png Size: 44638 bytes Desc: not available URL: From r.vandermeij at donders.ru.nl Sun Aug 26 01:06:28 2012 From: r.vandermeij at donders.ru.nl (Roemer van der Meij) Date: Sun, 26 Aug 2012 01:06:28 +0200 Subject: [FieldTrip] ft_databrowser In-Reply-To: References: Message-ID: Hi Ayelet, This is an intended feature, to make sure each trial is plotted at the same 'zoom-level' (keeping the plotted size of the time-axis the same). We hope this will help in recognizing odd patterns and such. You can also now zoom-in into the time-axis by pressing shift+left or right, or by using the buttons at the lower left corner. Hope this info helps, Best, Roemer On Sat, Aug 25, 2012 at 9:03 PM, Ayelet Landau wrote: > Hi all, > I thought I would report here also another (perhaps related?) strange > behavior that I encountered with databrowser. If you look at the attached > image you will immediately note that there is an empty portion after the > end of my data. The data plotted is indeed all the data (so, no missing > data), but this extra white just appears to varying degrees from trial to > trial. > > any idea why this occurs? > > thanks! > Ayelet > > [image: Inline image 1] > > On Sat, Aug 25, 2012 at 8:09 PM, Nenad Polomac wrote: > >> >> Dear Roemer, >> >> Thank you very much for your suggestion. Now works fine and much faster. >> Keep up the good work! :) >> >> I am sorry but I don't know how to put replay massage in the same thread >> in the Fieldtrip discussion list . Please give me some clue >> >> All the best! >> >> Nenad >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip >> > > > > -- > Ayelet N. Landau > > Postdoctoral Scientist, > Ernst Strüngmann Institute (ESI) for Neuroscience > in Cooperation with Max Planck Society > Deutschordenstr. 46, D-60528 Frankfurt > > Mobile: +49 (0)16 22733 110 > Fax: +49 (0)69 96769 555 > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- Roemer van der Meij M.Sc. PhD student Donders Institute for Brain, Cognition and Behaviour Centre for Cognition P.O. Box 9104 6500 HE Nijmegen The Netherlands Tel: +31(0)24 3655932 E-mail: r.vandermeij at donders.ru.nl -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.png Type: image/png Size: 44638 bytes Desc: not available URL: From karl.doron at gmail.com Sun Aug 26 20:53:53 2012 From: karl.doron at gmail.com (Karl Doron) Date: Sun, 26 Aug 2012 11:53:53 -0700 Subject: [FieldTrip] connectivity/amplitude correlation Message-ID: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> Hello, I'm wondering if there are any publications with the maths behind the functions for ft_connectivity_corr. I would specifically like to confirm (for a pending publication) how the amplitude correlation is computed when the function is given output from ft_freqanalysis which contains a time dimension (relevant cfg parameters pasted below). I've assumed this is occurring in the frequency domain across trials for sensor i and sensor j. cfg.channelcmb = chancmb; cfg.method = 'mtmconvol'; cfg.output = 'fourier'; cfg.keeptrials = 'yes'; cfg.foi = band(1):band(2); cfg.toi = 1.8:0.01:3.0; Thanks for any feedback! Karl Doron, Ph.D. UC, Santa Barbara -------------- next part -------------- An HTML attachment was scrubbed... URL: From stan.vanpelt at fcdonders.ru.nl Mon Aug 27 09:59:02 2012 From: stan.vanpelt at fcdonders.ru.nl (Stan van Pelt) Date: Mon, 27 Aug 2012 09:59:02 +0200 (CEST) Subject: [FieldTrip] ft_databrowser In-Reply-To: Message-ID: <1604964460.83554.1346054342785.JavaMail.root@sculptor.zimbra.ru.nl> Hi Roemer, Interesting features! In that respect, it might also be useful to set the x-axis ticklabels at 'logical'/more intuitive values (e.g., 0.1s apart) instead of 1/xth of the total data length. Best, Stan ----- Oorspronkelijk bericht ----- > Van: "Roemer van der Meij" > Aan: "ayelet landau" , "FieldTrip discussion > list" > Verzonden: Zondag 26 augustus 2012 01:06:28 > Onderwerp: Re: [FieldTrip] ft_databrowser > Hi Ayelet, > This is an intended feature, to make sure each trial is plotted at the > same 'zoom-level' (keeping the plotted size of the time-axis the > same). We hope this will help in recognizing odd patterns and such. > You can also now zoom-in into the time-axis by pressing shift+left or > right, or by using the buttons at the lower left corner. > Hope this info helps, > Best, > Roemer > On Sat, Aug 25, 2012 at 9:03 PM, Ayelet Landau < > ayelet.landau at gmail.com > wrote: > > Hi all, > > I thought I would report here also another (perhaps related?) > > strange > > behavior that I encountered with databrowser. If you look at the > > attached image you will immediately note that there is an empty > > portion after the end of my data. The data plotted is indeed all the > > data (so, no missing data), but this extra white just appears to > > varying degrees from trial to trial. > > any idea why this occurs? > > thanks! > > Ayelet > > Inline image 1 > > On Sat, Aug 25, 2012 at 8:09 PM, Nenad Polomac < > > polomacnenad at gmail.com > wrote: > > > Dear Roemer, > > > Thank you very much for your suggestion. Now works fine and much > > > faster. > > > Keep up the good work! :) > > > I am sorry but I don't know how to put replay massage in the same > > > thread in the Fieldtrip discussion list . Please give me some clue > > > All the best! > > > Nenad > > > _______________________________________________ > > > fieldtrip mailing list > > > fieldtrip at donders.ru.nl > > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > -- > > Ayelet N. Landau > > Postdoctoral Scientist, > > Ernst Strüngmann Institute (ESI) for Neuroscience > > in Cooperation with Max Planck Society > > Deutschordenstr. 46, D-60528 Frankfurt > > Mobile: +49 (0)16 22733 110 > > Fax: +49 (0)69 96769 555 > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- > Roemer van der Meij M.Sc. > PhD student > Donders Institute for Brain, Cognition and Behaviour > Centre for Cognition > P.O. Box 9104 > 6500 HE Nijmegen > The Netherlands > Tel: +31(0)24 3655932 > E-mail: r.vandermeij at donders.ru.nl > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Stan van Pelt Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen Kapittelweg 29, 6525 EN Nijmegen, Netherlands E-mail: stan.vanpelt at donders.ru.nl Website: www.ru.nl/donders/ Phone: (+31) (0)24 36 10981 Fax: (+31) (0)24 36 10989 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.png Type: image/png Size: 44638 bytes Desc: not available URL: From Lilla.Magyari at mpi.nl Mon Aug 27 11:59:53 2012 From: Lilla.Magyari at mpi.nl (Magyari, Lilla) Date: Mon, 27 Aug 2012 11:59:53 +0200 Subject: [FieldTrip] connectivity/amplitude correlation In-Reply-To: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> References: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> Message-ID: <503B4519.5030506@mpi.nl> Dear Karl, There are four papers mentioned in the help of ft_connectivity_corr. I am wondering if any of those could answer your question. Best, Lilla Karl Doron wrote: > Hello, > > I'm wondering if there are any publications with the maths behind the > functions for ft_connectivity_corr. I would specifically like to confirm > (for a pending publication) how the amplitude correlation is computed > when the function is given output from ft_freqanalysis which contains a > time dimension (relevant cfg parameters pasted below). I've assumed this > is occurring in the frequency domain across trials for sensor i and > sensor j. > > cfg.channelcmb = chancmb; > cfg.method = 'mtmconvol'; > cfg.output = 'fourier'; > cfg.keeptrials = 'yes'; > cfg.foi = band(1):band(2); > cfg.toi = 1.8:0.01:3.0; > > > Thanks for any feedback! > > > Karl Doron, Ph.D. > UC, Santa Barbara > > > > ------------------------------------------------------------------------ > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From karl.doron at gmail.com Mon Aug 27 16:16:52 2012 From: karl.doron at gmail.com (Karl Doron) Date: Mon, 27 Aug 2012 07:16:52 -0700 Subject: [FieldTrip] connectivity/amplitude correlation In-Reply-To: <503B4519.5030506@mpi.nl> References: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> <503B4519.5030506@mpi.nl> Message-ID: <0F563C8B-4EA4-43F2-9FF3-93625C9E4A1D@gmail.com> Hi Lilla, There is a reference for Rosenberg et al, 1998 for the partialization method in the code itself but I'm not seeing any in the help section. Thanks, karl On Aug 27, 2012, at 2:59 AM, "Magyari, Lilla" wrote: > Dear Karl, > > There are four papers mentioned in the help of ft_connectivity_corr. I am wondering if any of those could answer your question. > > Best, > Lilla > > Karl Doron wrote: >> Hello, >> I'm wondering if there are any publications with the maths behind the functions for ft_connectivity_corr. I would specifically like to confirm (for a pending publication) how the amplitude correlation is computed when the function is given output from ft_freqanalysis which contains a time dimension (relevant cfg parameters pasted below). I've assumed this is occurring in the frequency domain across trials for sensor i and sensor j. >> cfg.channelcmb = chancmb; >> cfg.method = 'mtmconvol'; >> cfg.output = 'fourier'; >> cfg.keeptrials = 'yes'; >> cfg.foi = band(1):band(2); >> cfg.toi = 1.8:0.01:3.0; >> Thanks for any feedback! >> Karl Doron, Ph.D. >> UC, Santa Barbara >> ------------------------------------------------------------------------ >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From Lilla.Magyari at mpi.nl Mon Aug 27 16:56:37 2012 From: Lilla.Magyari at mpi.nl (Magyari, Lilla) Date: Mon, 27 Aug 2012 16:56:37 +0200 Subject: [FieldTrip] connectivity/amplitude correlation In-Reply-To: <0F563C8B-4EA4-43F2-9FF3-93625C9E4A1D@gmail.com> References: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> <503B4519.5030506@mpi.nl> <0F563C8B-4EA4-43F2-9FF3-93625C9E4A1D@gmail.com> Message-ID: <503B8AA5.6020207@mpi.nl> hi Karl, Maybe your version of FieldTrip is older, and therefore, it does not contain the references yet. You can check the reference documentation of ft_connectivity_corr on the FieldTrip wiki: http://fieldtrip.fcdonders.nl/reference/ft_connectivity_corr (The reference documentation of the functions is always automatically updated according to the latest version of the documentation part (or "help" part) of the script.) The same articles (and more) can be found also under Literature (Method References) http://fieldtrip.fcdonders.nl/references_to_implemented_methods under the title Connectivity Analysis. I have not read the articles myself, so I am not sure if they give the information you need. But I was just wondering if you are aware of the recently available documentation of the function. Best, Lilla Karl Doron wrote: > Hi Lilla, > > There is a reference for Rosenberg et al, 1998 for the partialization method in the code itself but I'm not seeing any in the help section. > > Thanks, > karl > > > On Aug 27, 2012, at 2:59 AM, "Magyari, Lilla" wrote: > >> Dear Karl, >> >> There are four papers mentioned in the help of ft_connectivity_corr. I am wondering if any of those could answer your question. >> >> Best, >> Lilla >> >> Karl Doron wrote: >>> Hello, >>> I'm wondering if there are any publications with the maths behind the functions for ft_connectivity_corr. I would specifically like to confirm (for a pending publication) how the amplitude correlation is computed when the function is given output from ft_freqanalysis which contains a time dimension (relevant cfg parameters pasted below). I've assumed this is occurring in the frequency domain across trials for sensor i and sensor j. >>> cfg.channelcmb = chancmb; >>> cfg.method = 'mtmconvol'; >>> cfg.output = 'fourier'; >>> cfg.keeptrials = 'yes'; >>> cfg.foi = band(1):band(2); >>> cfg.toi = 1.8:0.01:3.0; >>> Thanks for any feedback! >>> Karl Doron, Ph.D. >>> UC, Santa Barbara >>> ------------------------------------------------------------------------ >>> _______________________________________________ >>> fieldtrip mailing list >>> fieldtrip at donders.ru.nl >>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From r.oostenveld at donders.ru.nl Tue Aug 28 10:05:25 2012 From: r.oostenveld at donders.ru.nl (Robert Oostenveld) Date: Tue, 28 Aug 2012 10:05:25 +0200 Subject: [FieldTrip] PLV on sources In-Reply-To: <201208241235.q7OCZhaZ002930@brievenbus.science.ru.nl> References: <201208241235.q7OCZhaZ002930@brievenbus.science.ru.nl> Message-ID: Dear Robert and Nayaran, The data handling in ft_connectivityanalysis is not yet at the level that it can easily deal with all source representations; that is something that we will improve upon in the future. The way we suggest it now is similar to your suggestion and is outlined in detail in the http://fieldtrip.fcdonders.nl/tutorial/connectivity tutorial. Specifically have a look at the second section where as example source (virtual channel) coherence and granger is computed. In that tutorial you would just change cfg.method in plv and you should be rolling. Best regards, Robert O. PS this connectivity tutorial is quite new, so might also be of interest for other people doing connectivity analysis! Op 24 aug. 2012 om 14:35 heeft "Narayan Puthanmadam Subramaniyam" het volgende geschreven: > hi > > i was thinking of doing the same thing exactly. Computing plv on the eeg inverse solution. technically it should be possible going by the general level concept of phase locking. but i am not sure so let the experts here speak. > > Thanks&Regards, > NPS > > Sent from my Nokia phone > -----Original Message----- > From: Robert Brown > Sent: 24:08:2012, 11:38 > To: fieldtrip at science.ru.nl > Subject: [FieldTrip] PLV on sources > > > Dear Fieldtrippers, > > I see (from the ft_connectivityanalysis code) that one cannot run phase > locking (PLV) analysis (a la Lachaux et al) on source data. > > My brain still insisted on trying to do so but unfortunately I am not sure > if it is clever enough. Hope you can help! > > It seems that a way to go would be to do LCMV beamforming on the data, get > the virtual electrodes, then run freqanalysis with fourier output on these > virtual electrodes and then perform PLV on those data. > > Before I dig myself very deep into this, maybe some experts here could > guide me: Would this approach work? Any possible caveats? Are there maybe > other, better and more straightforward ways for achieving the PLV on source > level? > > Any comments would be appreciated. Thank you very much for your time! > > Kind regards, > Bob > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From sherrykhan78 at gmail.com Tue Aug 28 12:34:14 2012 From: sherrykhan78 at gmail.com (Sheraz Khan) Date: Tue, 28 Aug 2012 06:34:14 -0400 Subject: [FieldTrip] PLV on sources In-Reply-To: References: <201208241235.q7OCZhaZ002930@brievenbus.science.ru.nl> Message-ID: Dear Bob, Another possible solution is to do wavelet decomposition in sensor space and then multiply it with your imaging kernel, then do PLV in source space, this will save some computational time, but still you have to deal with EEG/MEG point spread function. Sheraz Khan Martinos Center MGH/MIT/Harvard On Tue, Aug 28, 2012 at 4:05 AM, Robert Oostenveld < r.oostenveld at donders.ru.nl> wrote: > Dear Robert and Nayaran, > > The data handling in ft_connectivityanalysis is not yet at the level that > it can easily deal with all source representations; that is something that > we will improve upon in the future. > > The way we suggest it now is similar to your suggestion and is outlined in > detail in the http://fieldtrip.fcdonders.nl/tutorial/connectivitytutorial. Specifically have a look at the second section where as example > source (virtual channel) coherence and granger is computed. In that > tutorial you would just change cfg.method in plv and you should be rolling. > > Best regards, > Robert O. > > PS this connectivity tutorial is quite new, so might also be of interest > for other people doing connectivity analysis! > > > Op 24 aug. 2012 om 14:35 heeft "Narayan Puthanmadam Subramaniyam" < > narayan.ps at tut.fi> het volgende geschreven: > > > hi > > > > i was thinking of doing the same thing exactly. Computing plv on the eeg > inverse solution. technically it should be possible going by the general > level concept of phase locking. but i am not sure so let the experts here > speak. > > > > Thanks&Regards, > > NPS > > > > Sent from my Nokia phone > > -----Original Message----- > > From: Robert Brown > > Sent: 24:08:2012, 11:38 > > To: fieldtrip at science.ru.nl > > Subject: [FieldTrip] PLV on sources > > > > > > Dear Fieldtrippers, > > > > I see (from the ft_connectivityanalysis code) that one cannot run phase > > locking (PLV) analysis (a la Lachaux et al) on source data. > > > > My brain still insisted on trying to do so but unfortunately I am not > sure > > if it is clever enough. Hope you can help! > > > > It seems that a way to go would be to do LCMV beamforming on the data, > get > > the virtual electrodes, then run freqanalysis with fourier output on > these > > virtual electrodes and then perform PLV on those data. > > > > Before I dig myself very deep into this, maybe some experts here could > > guide me: Would this approach work? Any possible caveats? Are there maybe > > other, better and more straightforward ways for achieving the PLV on > source > > level? > > > > Any comments would be appreciated. Thank you very much for your time! > > > > Kind regards, > > Bob > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -------------- next part -------------- An HTML attachment was scrubbed... URL: From r.oostenveld at donders.ru.nl Thu Aug 30 10:31:17 2012 From: r.oostenveld at donders.ru.nl (Robert Oostenveld) Date: Thu, 30 Aug 2012 10:31:17 +0200 Subject: [FieldTrip] PLV on sources In-Reply-To: References: <201208241235.q7OCZhaZ002930@brievenbus.science.ru.nl> Message-ID: Dear Sheraz, Bob You are completely right; since the fft and the spatial filter multiplication are both linear operations, their order can be interchanged. If nsource>nchan, this will speed things up. For full brain source reconstructions it will be much faster (and require less memory). It would be nice to extend the tutorial with a section that explains precisely this. Best Robert Op 28 aug. 2012 om 12:34 heeft Sheraz Khan het volgende geschreven: > Dear Bob, > > Another possible solution is to do wavelet decomposition in sensor space and then multiply it with your imaging kernel, then do PLV in source space, this will save some computational time, but still you have to deal with EEG/MEG point spread function. > > Sheraz Khan > Martinos Center > MGH/MIT/Harvard > > On Tue, Aug 28, 2012 at 4:05 AM, Robert Oostenveld wrote: > Dear Robert and Nayaran, > > The data handling in ft_connectivityanalysis is not yet at the level that it can easily deal with all source representations; that is something that we will improve upon in the future. > > The way we suggest it now is similar to your suggestion and is outlined in detail in the http://fieldtrip.fcdonders.nl/tutorial/connectivity tutorial. Specifically have a look at the second section where as example source (virtual channel) coherence and granger is computed. In that tutorial you would just change cfg.method in plv and you should be rolling. > > Best regards, > Robert O. > > PS this connectivity tutorial is quite new, so might also be of interest for other people doing connectivity analysis! > > > Op 24 aug. 2012 om 14:35 heeft "Narayan Puthanmadam Subramaniyam" het volgende geschreven: > > > hi > > > > i was thinking of doing the same thing exactly. Computing plv on the eeg inverse solution. technically it should be possible going by the general level concept of phase locking. but i am not sure so let the experts here speak. > > > > Thanks&Regards, > > NPS > > > > Sent from my Nokia phone > > -----Original Message----- > > From: Robert Brown > > Sent: 24:08:2012, 11:38 > > To: fieldtrip at science.ru.nl > > Subject: [FieldTrip] PLV on sources > > > > > > Dear Fieldtrippers, > > > > I see (from the ft_connectivityanalysis code) that one cannot run phase > > locking (PLV) analysis (a la Lachaux et al) on source data. > > > > My brain still insisted on trying to do so but unfortunately I am not sure > > if it is clever enough. Hope you can help! > > > > It seems that a way to go would be to do LCMV beamforming on the data, get > > the virtual electrodes, then run freqanalysis with fourier output on these > > virtual electrodes and then perform PLV on those data. > > > > Before I dig myself very deep into this, maybe some experts here could > > guide me: Would this approach work? Any possible caveats? Are there maybe > > other, better and more straightforward ways for achieving the PLV on source > > level? > > > > Any comments would be appreciated. Thank you very much for your time! > > > > Kind regards, > > Bob > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Thu Aug 30 10:33:28 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Thu, 30 Aug 2012 10:33:28 +0200 Subject: [FieldTrip] neighbours Message-ID: Hi, I am analyzing data obtained on CTF MEG 275 system. However, 4 sensors are broken and data ended up with 271 sensor. So, my question is should I interpolate the missing sensors with FT_CHANNELREPAI *R*? In the case I leave data with 271 sensor, will those missing sensors influence neighbors configuration in the statistic and the source localisation analysis? Thank you in advance! Kind regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 30 21:57:16 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 30 Aug 2012 21:57:16 +0200 Subject: [FieldTrip] connectivity/amplitude correlation In-Reply-To: <0F563C8B-4EA4-43F2-9FF3-93625C9E4A1D@gmail.com> References: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> <503B4519.5030506@mpi.nl> <0F563C8B-4EA4-43F2-9FF3-93625C9E4A1D@gmail.com> Message-ID: <274B2F2C-FE26-4A89-A2EC-84405BB77C2F@donders.ru.nl> Dear Karl, Perhaps I can add one or two things to Lilla's reply. If I understand your question correctly, it pertains specifically to the math behind the amplitude correlation computation, and how this is done in ft_connectivity_corr. It is indeed as you thought, the dimension over which the correlation is computed is indeed the trial dimension. That is, for each channelpair-time-frequency-point the amplitude correlation is computed across trials. A lot of the intelligence already takes place in ft_connectivityanalysis. In your case ft_connectivityanalysis computes the covariance (and variance) across trials between the amplitudes. The only thing that happens in ft_connectivity_corr is the normalisation with the square root of the variance product. Hope this helps, Jan-Mathijs On Aug 27, 2012, at 4:16 PM, Karl Doron wrote: > Hi Lilla, > > There is a reference for Rosenberg et al, 1998 for the partialization method in the code itself but I'm not seeing any in the help section. > > Thanks, > karl > > > On Aug 27, 2012, at 2:59 AM, "Magyari, Lilla" wrote: > >> Dear Karl, >> >> There are four papers mentioned in the help of ft_connectivity_corr. I am wondering if any of those could answer your question. >> >> Best, >> Lilla >> >> Karl Doron wrote: >>> Hello, >>> I'm wondering if there are any publications with the maths behind the functions for ft_connectivity_corr. I would specifically like to confirm (for a pending publication) how the amplitude correlation is computed when the function is given output from ft_freqanalysis which contains a time dimension (relevant cfg parameters pasted below). I've assumed this is occurring in the frequency domain across trials for sensor i and sensor j. >>> cfg.channelcmb = chancmb; >>> cfg.method = 'mtmconvol'; >>> cfg.output = 'fourier'; >>> cfg.keeptrials = 'yes'; >>> cfg.foi = band(1):band(2); >>> cfg.toi = 1.8:0.01:3.0; >>> Thanks for any feedback! >>> Karl Doron, Ph.D. >>> UC, Santa Barbara >>> ------------------------------------------------------------------------ >>> _______________________________________________ >>> fieldtrip mailing list >>> fieldtrip at donders.ru.nl >>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jm.horschig at donders.ru.nl Fri Aug 31 09:50:17 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Fri, 31 Aug 2012 09:50:17 +0200 Subject: [FieldTrip] neighbours In-Reply-To: References: Message-ID: <50406CB9.7090209@donders.ru.nl> Hi Nenad, the question you are asking does not deserve a definite 'yes' or 'no'. In any case, note that when averaging over subjects, missing sensors for one subject will be removed for all other subjects as well. So, if you are talking about different missing sensors per subject, an interpolation will definitely be a wise thing to do. In case you choose for interpolating the missing channels, you can try to check out the new spherical spline interpolation method (see help ft_channelrepair) - theoretically that should result in a quite good reconstruction. But the (standard) nearest neighbour interpolation is fine as well. With respect to your second question: missing sensors will influence the neighbour configuration depending on the method you chose for neighbour selection. For the 'distance' and 'template' method, each neighbouring channel of the missing will have one sensor less (i.e. the missing one). The 'triangulation' method will create triangle neighbour no matter how many channels you have, so it will just ignore the whole in space. In any case, I'd propose that you check out how satisfied you personally are with the neighbours by calling ft_neighbourplot. Just last week, I added the new option that if you use cfg.enableedit='yes', you can interactively change the neighbourhood structure within the neighbourplot. If I were you, I would start from the template method. As an additional piece of information: I am currently working on improving the neighbour templates, and will upload them probably next week. All that advertised, let me also say that having 4 out of 275 sensors missing won't affect your statistics much. Here, it of course depends on your region if interest, research question etc. E.g. if you are interested in posterior alpha power, but miss four temporal channels, there is no reason to worry. So, without knowing what you are looking for and where the missing channels are, I cannot give you a general advise what to do. As a last remark, the sensitiviy of the statistics will of course be influenced by missing sensors, but I doubt that it will matter much. Btw, I am not quite sure how much the spherical spline interpolation will buy you here, you might give it a try. I bet that it will increase your sensitivity more than the nearest neighbour approach, although both might result in rather low increases. But this is just a general gut feeling rather than anything I empirically validated. Best, Jörn On 8/30/2012 10:33 AM, Nenad Polomac wrote: > Hi, > > I am analyzing data obtained on CTF MEG 275 system. However, 4 sensors > are broken and data ended up with 271 sensor. So, my question is > should I interpolate the missing sensors with FT_CHANNELREPAI > _R_? In the > case I leave data with 271 sensor, will those missing sensors > influence neighbors configuration in the statistic and the source > localisation analysis? > > Thank you in advance! > > Kind regards! > > Nenad > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group FieldTrip Development Team P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From akiko.ikkai at gmail.com Fri Aug 31 21:59:00 2012 From: akiko.ikkai at gmail.com (Akiko Ikkai) Date: Fri, 31 Aug 2012 15:59:00 -0400 Subject: [FieldTrip] FT_SCALPCURRENTDENSITY options and elec units Message-ID: Hi, I'm trying to run plv analysis, first acquiring SCD of my data with ft_scalcurrentdensity, and am confused about the difference between cfg.method = 'finite' and 'spline.' Both gave similar results short-range, but 'spline' gives higher plv at long range plv. I'd love it if someone could explain or suggest reading on the difference between these two methods (or is one method preferred over the other in some cases?). Also, I'm not sure whether or not I should convert my elec.pnt values, since the code ft_scalcurrentdensity says that "Note that the skin conductivity, electrode dimensions and the potential all have to be expressed in the same SI units." My elec.pnt are in cm, and I'm using the default conductivity (0.33 S/m). Do I need to do something like elec.pnt = elec.pnt/100 to convert the unit from cm to m? With or without conversion, I got the same results (viewed with ft_topoplotTFR), and I'm a little confused... Thank you in advance for your time! Akiko -- Akiko Ikkai, Ph.D. Postdoctoral Fellow Department of Psychological and Brain Sciences Johns Hopkins University Ames Hall, 3400 N. Charles St. Baltimore, MD 21218 -------------- next part -------------- An HTML attachment was scrubbed... URL: From dohuyn2001 at hotmail.com Wed Aug 1 07:17:37 2012 From: dohuyn2001 at hotmail.com (DoHyun Kim) Date: Wed, 1 Aug 2012 05:17:37 +0000 Subject: [FieldTrip] Question about BCI2000 2D cursor task with Field trip buffer Message-ID: Hi, I'm a college student currently working on the BCI2000 2D cursor task with Field trip buffer. To moves cursors all directions (up, down, left, and right), I followed the description from the wiki that event_up = event; event_up.value = 1; event_down = event; event_down.value = -1; event_null = event; event_null.value = 0; (see http://www.bci2000.org/wiki/index.php/Programming_Tutorial:Working_with_the_FieldTrip_buffer) Currently, my cursor moves right and left with settings as event_right.value=1, event_left.value =-1. While I'm trying to move my cursor on y-axis(since I'm aiming for 2D control), I couldn't figure out how I should write the script for up & down movements. Does anyone know how to do it? Thanks. -------------- next part -------------- An HTML attachment was scrubbed... URL: From vitoria.piai at gmail.com Wed Aug 1 13:22:49 2012 From: vitoria.piai at gmail.com (=?ISO-8859-1?Q?Vit=F3ria_Magalh=E3es_Piai?=) Date: Wed, 01 Aug 2012 13:22:49 +0200 Subject: [FieldTrip] plotting mri ortho with crosshair Message-ID: <50191189.6080809@gmail.com> Hi all, I'm trying to plot an MR image (not necessarily functional data) with the ortho method and with crosshairs but the crosshairs are not there at all. I used to be able to see them before and with interactive mode I was able to go through the coordinates and still seeing where the crosshairs were. What I need in the end is to be able to see on an MRI (without functional data) with crosshairs where exactly a certain coordinate is. Any suggestions? If I remember correctly, I used to plot it using: mri = ft_read_mri(template); cfg.interactive = 'yes'; cfg.crosshair = 'yes'; ft_sourceplot(cfg, mri) Thanx a lot, Vitória -- ** Please consider the environment - do you really need to print? ** From jan.schoffelen at donders.ru.nl Wed Aug 1 13:41:41 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Wed, 1 Aug 2012 13:41:41 +0200 Subject: [FieldTrip] plotting mri ortho with crosshair In-Reply-To: <50191189.6080809@gmail.com> References: <50191189.6080809@gmail.com> Message-ID: Hi Vitoria, Problem confirmed. Could you file this as a bug on bugzilla? We can take it from there. Best, Jan-Mathijs On Aug 1, 2012, at 1:22 PM, Vitória Magalhães Piai wrote: > Hi all, > > I'm trying to plot an MR image (not necessarily functional data) with the ortho method and with crosshairs but the crosshairs are not there at all. > I used to be able to see them before and with interactive mode I was able to go through the coordinates and still seeing where the crosshairs were. > What I need in the end is to be able to see on an MRI (without functional data) with crosshairs where exactly a certain coordinate is. > Any suggestions? > > If I remember correctly, I used to plot it using: > > mri = ft_read_mri(template); > cfg.interactive = 'yes'; > cfg.crosshair = 'yes'; > ft_sourceplot(cfg, mri) > > Thanx a lot, Vitória > > -- > ** Please consider the environment - do you really need to print? ** > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Wed Aug 1 15:00:13 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Wed, 1 Aug 2012 15:00:13 +0200 Subject: [FieldTrip] plotting mri ortho with crosshair In-Reply-To: <50191189.6080809@gmail.com> References: <50191189.6080809@gmail.com> Message-ID: Hi Vitoria, After consulting with Eelke, it seems that the problem is caused by some low-level bug in the graphics handling. More to the point, there seems to be something going on with the openGL rendering in some matlab versions. If you switch to cfg.renderer = 'zbuffer' you should be able to see the cross hair (but you will loose the opacity masking for functional data). For now this is the best we can offer. Cheers, JM On Aug 1, 2012, at 1:22 PM, Vitória Magalhães Piai wrote: > Hi all, > > I'm trying to plot an MR image (not necessarily functional data) with the ortho method and with crosshairs but the crosshairs are not there at all. > I used to be able to see them before and with interactive mode I was able to go through the coordinates and still seeing where the crosshairs were. > What I need in the end is to be able to see on an MRI (without functional data) with crosshairs where exactly a certain coordinate is. > Any suggestions? > > If I remember correctly, I used to plot it using: > > mri = ft_read_mri(template); > cfg.interactive = 'yes'; > cfg.crosshair = 'yes'; > ft_sourceplot(cfg, mri) > > Thanx a lot, Vitória > > -- > ** Please consider the environment - do you really need to print? ** > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Wed Aug 1 17:55:13 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Wed, 1 Aug 2012 17:55:13 +0200 Subject: [FieldTrip] prepare neighbors template file Message-ID: Hi, I am analyzing CTF MEG data with 275 channels. I would like to know where can I find neighbors template file. Or should I create one by myself! I need this file for planar gradient calculation. Thank you in advance! All the best! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From marco.buiatti at gmail.com Wed Aug 1 18:35:53 2012 From: marco.buiatti at gmail.com (Marco Buiatti) Date: Wed, 1 Aug 2012 18:35:53 +0200 Subject: [FieldTrip] interactions between two factors In-Reply-To: References: Message-ID: Thanks Arno and Stephen for your replies. Stephen, the message you cite confirms that what I propose in my first point is correct, thanks. Any statistician willing to answer to my points 2) and 3) is welcome. Best, Marco On 27 July 2012 16:25, Stephen Politzer-Ahles wrote: > Hello Marco, > > For how to test an interaction in a 2x2 design, see > http://mailman.science.ru.nl/pipermail/fieldtrip/2011-January/003447.html > (and some additional information at > http://mailman.science.ru.nl/pipermail/fieldtrip/2010-December/003338.html, > http://mailman.science.ru.nl/pipermail/fieldtrip/2011-September/004244.html, > and Anderson & Braak 2003 in Journal of Statistical Computation & > Simulation). I'm not sure about the other two issues you mentioned, though. > > Best, > Steve > >> Message: 1 >> Date: Thu, 26 Jul 2012 12:36:39 +0200 >> From: Marco Buiatti >> To: fieldtrip at donders.ru.nl >> Subject: Re: [FieldTrip] interactions between two factors >> Message-ID: >> >> >> Content-Type: text/plain; charset=ISO-8859-1 >> >> >> Dear FieldTrippers, >> >> some time ago I have posted the message below concerning how to >> compute the statistical interaction between two factors in an EEG >> study with the FieldTrip cluster-based statistical analysis. Since I >> believe it is a problem of general interest, I was confident I would >> have received some replies, no matter how critics. But I had no reply >> and I am trying to guess why: >> >> - the problem is trivial, I should go back to my statistics books and >> solve it myself; >> - the problem is ill-posed, I should go back to my statistics books >> and reformulate it correctly; >> - the problem is tabou, no one dares commits to a solution because it >> could be a wrong one. >> - the problem is solved: I should read message number #. >> >> Thanks a lot for your feedback, >> >> Best, >> >> Marco >> >> On 25 May 2012 15:58, Marco Buiatti wrote: >> > Dear FieldTrippers, >> > >> > I am analysing an EEG study with 2x4 factors: one varies between 4 >> > parametrically varying levels (1 to 4), the second between two levels. >> > >> > I have three questions concerning the use of Fieldtrip cluster-based >> > non parametric statistical analysis in this case: >> > >> > 1) How to compute the interaction between the two factors. Let's start >> > from the simplest case of a 2x2 design, factors varying between values >> > A1 and A2 for the first factor, B1 and B2 for the second. Please tell >> > me if it is correct to compute the interaction by: >> > - computing the difference diffA=ERP(A1)-ERP(A2) separately in >> > condition B1 and B2, for every subject >> > - performing a within-subjects statistical analysis between diffA in >> > condition B1 and diffA in condition B2 (function >> > statfun_depsamplesT.m). >> > >> > 2) Now consider that factor A varies parametrically between values 1 >> > to 4. For the main effect of this factor, I have used the Fieldtrip >> > function statfun_depsamplesregrT.m and I'm satisfied with it. Is it >> > correct to compute the interaction by >> > - computing the regression >> > regrA=regression(ERP(A1),ERP(A2),ERP(A3),ERP(A4)) (computed as inside >> > function statfun_depsamplesregrT.m) separately in condition B1 and B2, >> > for every subject >> > - performing a within-subjects statistical analysis between regrA in >> > condition B1 and regrA in condition B2 (function >> > statfun_depsamplesT.m)? >> > >> > 3) Since BEFORE looking at the data (this is to prevent Eric's >> > contestation...) I expect a dipolar topography for the regression >> > (data are in average reference), I would like to combine into a joint >> > cluster negative and positive clusters. I have tried by changing >> > statfun_depsamplesregrT.m by just taking the absolute value of the >> > regression, but I get weird results (a huge, non significant cluster). >> > Is it possible that since values are now all positive, I should use a >> > different statistical test at the single bin level? Any other >> > suggestions? >> > >> > Thanks in advance for your help, >> > >> > Marco >> > >> > >> > >> > -- >> > Marco Buiatti, PhD >> > >> > CEA/DSV/I2BM / NeuroSpin >> > INSERM U992 - Cognitive Neuroimaging Unit >> > B?t 145 - Point Courrier 156 >> >> > Gif sur Yvette F-91191 FRANCE >> > Ph: +33(0)169.08.65.21 >> > Fax: +33(0)169.08.79.73 >> > E-mail: marco.buiatti at gmail.com >> > http://www.unicog.org/pm/pmwiki.php/Main/MarcoBuiatti >> > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Marco Buiatti, PhD CEA/DSV/I2BM / NeuroSpin INSERM U992 - Cognitive Neuroimaging Unit Bât 145 - Point Courrier 156 Gif sur Yvette F-91191 FRANCE Ph: +33(0)169.08.65.21 Fax: +33(0)169.08.79.73 E-mail: marco.buiatti at gmail.com http://www.unicog.org/pm/pmwiki.php/Main/MarcoBuiatti *********************************************** From polomacnenad at gmail.com Wed Aug 1 18:41:56 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Wed, 1 Aug 2012 18:41:56 +0200 Subject: [FieldTrip] MEG-anatomical MRI image coregistration Message-ID: Dear Stephen, Thank you very much for your expertise! All the best! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From lihqih at gmail.com Wed Aug 1 19:33:38 2012 From: lihqih at gmail.com (qi li) Date: Wed, 1 Aug 2012 13:33:38 -0400 Subject: [FieldTrip] mne source reconstruction Message-ID: Hi, Is there anyway to map the source space mesh back to voxel space so I can use the ft_sourceplot. After MNE source reconstruction, I have source = time: [1x399 double] pos: [8196x3 double] inside: [1x8196 double] outside: [1x0 double] method: 'average' avg: [1x1 struct] cfg: [1x1 struct] I have 8196 mesh points but how to map it back to the anatomical voxel space? Thanks! Qi From jm.horschig at donders.ru.nl Thu Aug 2 08:49:06 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Thu, 02 Aug 2012 08:49:06 +0200 Subject: [FieldTrip] prepare neighbors template file In-Reply-To: References: Message-ID: <501A22E2.6020505@donders.ru.nl> Dear Nenad, the file can be found in fieldtrip/templates/neighbours It is sufficient to specify cfg.template = 'ctf275_neighb'; or cfg.layout = 'ctf275.lay'; Best, Jörn On 8/1/2012 5:55 PM, Nenad Polomac wrote: > Hi, > I am analyzing CTF MEG data with 275 channels. I would like to know > where can I find neighbors template file. Or should I create one by > myself! I need this file for planar gradient calculation. > > Thank you in advance! > > All the best! > > Nenad > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From jm.horschig at donders.ru.nl Thu Aug 2 08:52:46 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Thu, 02 Aug 2012 08:52:46 +0200 Subject: [FieldTrip] mne source reconstruction In-Reply-To: References: Message-ID: <501A23BE.8040706@donders.ru.nl> Dear Qi, Not quite sure if I understand your request correctly, but you can just rehape: / source.avg.pow = reshape(source.avg.pow, anatomical.dim)/; Then instead of being a vector, it will become a matrix. If that's not the answer you intended to get, please phrase your question differently :) Best, Jörn On 8/1/2012 7:33 PM, qi li wrote: > Hi, > > Is there anyway to map the source space mesh back to voxel space so I > can use the ft_sourceplot. > > After MNE source reconstruction, I have > > source = > > time: [1x399 double] > pos: [8196x3 double] > inside: [1x8196 double] > outside: [1x0 double] > method: 'average' > avg: [1x1 struct] > cfg: [1x1 struct] > > I have 8196 mesh points but how to map it back to the anatomical voxel > space? Thanks! > > Qi > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 2 08:59:04 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 2 Aug 2012 08:59:04 +0200 Subject: [FieldTrip] mne source reconstruction In-Reply-To: References: Message-ID: <5A868F5B-3DD0-4383-BF64-8D53B537A128@donders.ru.nl> Hi Qi, You might want to check the function 'ft_sourceinterpolate' for this. Best, Jan-Mathijs On Aug 1, 2012, at 7:33 PM, qi li wrote: > Hi, > > Is there anyway to map the source space mesh back to voxel space so I > can use the ft_sourceplot. > > After MNE source reconstruction, I have > > source = > > time: [1x399 double] > pos: [8196x3 double] > inside: [1x8196 double] > outside: [1x0 double] > method: 'average' > avg: [1x1 struct] > cfg: [1x1 struct] > > I have 8196 mesh points but how to map it back to the anatomical voxel > space? Thanks! > > Qi > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 2 09:07:22 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 2 Aug 2012 09:07:22 +0200 Subject: [FieldTrip] mne source reconstruction In-Reply-To: <501A23BE.8040706@donders.ru.nl> References: <501A23BE.8040706@donders.ru.nl> Message-ID: <4570E04B-810A-4002-B286-561FDE5C0221@donders.ru.nl> Hi Qi, To follow up on Jörn: If you reconstructed the source activations on a mesh that represents the cortical surface, you first need to interpolate the functional data into a 3D volume, if you want to use ft_sourceplot for visualization. For this you need ft_sourceinterpolate. As an alternative, to visualize the results, you can use ft_sourcemovie. However, this function is under development so it may not give the most visually appealing results. Best, JM On Aug 2, 2012, at 8:52 AM, Jörn M. Horschig wrote: > Dear Qi, > > Not quite sure if I understand your request correctly, but you can just rehape: > source.avg.pow = reshape(source.avg.pow, anatomical.dim); > Then instead of being a vector, it will become a matrix. > > If that's not the answer you intended to get, please phrase your question differently :) > > Best, > Jörn > > On 8/1/2012 7:33 PM, qi li wrote: >> Hi, >> >> Is there anyway to map the source space mesh back to voxel space so I >> can use the ft_sourceplot. >> >> After MNE source reconstruction, I have >> >> source = >> >> time: [1x399 double] >> pos: [8196x3 double] >> inside: [1x8196 double] >> outside: [1x0 double] >> method: 'average' >> avg: [1x1 struct] >> cfg: [1x1 struct] >> >> I have 8196 mesh points but how to map it back to the anatomical voxel >> space? Thanks! >> >> Qi >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > -- > Jörn M. Horschig > PhD Student > Donders Institute for Brain, Cognition and Behaviour > Centre for Cognitive Neuroimaging > Radboud University Nijmegen > Neuronal Oscillations Group > > P.O. Box 9101 > NL-6500 HB Nijmegen > The Netherlands > > Contact: > E-Mail: jm.horschig at donders.ru.nl > Tel: +31-(0)24-36-68493 > Web: http://www.ru.nl/donders > > Visiting address: > Trigon, room 2.30 > Kapittelweg 29 > NL-6525 EN Nijmegen > The Netherlands > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 2 09:16:26 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 2 Aug 2012 09:16:26 +0200 Subject: [FieldTrip] granger causality In-Reply-To: <20120730162938.56630@gmx.net> References: <20120730162938.56630@gmx.net> Message-ID: <17A8F077-820B-48A6-A388-2B2E0B7AB002@donders.ru.nl> Hi Carina, > 1.) I know that I can calculate the appropriate model order and window length with the bsmart toolbox. But from the documentation I somehow do not get the format of data I should use for this calculation. > When I enter the time-frequency data (which I want to use to estimate granger causality) (size: 2 x 2 x freq x time points), it doesn't work. The 'it doesn't work' statement is typically not enough for us to diagnose where the problem lies (at least at which step it goes wrong). At the moment I can only assume that you use an outdated version of FieldTrip, because when I try to do TF-resolved Granger using AR-models it works. It is good that you pasted the script below, but an error message may be informative too. > Any advise? I have a sampling rate of 1000. What sort of time window and maximal model order would make sense do you think? I think that the time window is rather short: I'd rather go for 0.4 or 0.5 to begin with, but this is an empirical question. Also, shifting the windows one time step at a time does not really make sense to me, 0.01 or even 0.05 is much more memory friendly. For the model order: higher one. Some people seem to get nice results when taking a relatively high model order. Of course, for a high model order you also need longer time-windows. > 2)I would like to keep the time domain analyzing my data. What do you think would be an appropriate sliding time window using the 'ft_mvaranalysis' function? And what does it mean exactly? By choosing the model order, I am already determining the maximal time lag between the two functions. Are the values then estimated for the whole time window? No. The time window determines the length of the data segment which is used to fit an AR-model (of order X) of the data at time point Y (as defined in your toi) > 3)What do you think about statistics? Would it makes sense to use the non-parametric cluster approach to shuffle within patients and do a group analysis that way? This depends on your experimental question and you null-hypothesis. Best, Jan-Mathijs > > Thank you so much in advance! > Best, > Carina > > > As a summary,I am doing following steps with field trip: > > > cfg = []; > cfg.dftfilter ='yes' > prep_cond1{subj} = ft_preprocessing(cfg, data); > > > cfg = []; > cfg.order = 5; > cfg.toolbox = 'bsmart'; > cfg.t_ftimwin = 0.05 > cfg.toi = -1:0.001:3.5; > mdata_cond1{subj}= ft_mvaranalysis(cfg, prep_cond1{subj}); > > > cfg = []; > cfg.method = 'mvar'; > cfg.foi = 4:100; > cond1_freq{subj}= ft_freqanalysis_mvar(cfg,mdata_cond1{subj}); > > > cfg = []; > cfg.method = 'granger'; > cond1_granger{subj} = ft_connectivityanalysis(cfg, cond1_freq{subj}); > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 2 09:27:38 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 2 Aug 2012 09:27:38 +0200 Subject: [FieldTrip] Question regarding ft_connectivitysimulation In-Reply-To: <5A1787011651BC42A4D41856DBC2E0603E048720@mbox-f-3.du.deakin.edu.au> References: <5A1787011651BC42A4D41856DBC2E0603E048720@mbox-f-3.du.deakin.edu.au> Message-ID: <6BC78C56-C12D-485A-B916-170A6D0EFCD7@donders.ru.nl> Hi Imali, I am not sure whether I understand completely what you want. I guess you want to simulate time series of 'sources', that have different (more or less well defined) spectral characteristics. As of yet, ft_connectivitysimulation does not support this. You could however call ft_connectivitysimulation twice, (using cfg.method = 'inear_mix') and using different cfg.bpfreq each time. Subsequently, you could sum the resulting time courses: for k = 1:numel(data1.trial) data1.trial{k} = data1.trial{k} + data2.trial{k}; end where data1 and data2 are the output to ft_connectivitysimulation. Best, Jan-Mathijs On Jul 19, 2012, at 6:44 AM, IMALI THANUJA HETTIARACHCHI wrote: > Dear fieldTrippers, > > I am trying to generate some simulated signals from a known connectivity structure with ft_connectivitysimulation, and secondly to assign these signals to known dipole locations when using in ft_dipolesimulation. > > Is there any way I can assign a frequency to the signals in ft_connectivityanalysis, so that I can assign different known frequency signals for dipoles? > > Any help is really appreciated. > > Thank you very much. > > Kind regards > Imali > > Imali Thanuja Hettiarachchi > PhD Candidate > Centre for Intelligent Systems research > Deakin University, Geelong 3217, Australia. > Email: ith at deakin.edu.au > www.deakin.edu.au/cisr > > > > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 2 09:31:03 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 2 Aug 2012 09:31:03 +0200 Subject: [FieldTrip] About co-register the individual sudbject MRI and sensor locations In-Reply-To: <5D2CC5653262C14188174E20065A4C4EC8D457@hscmbx6.uthsc.tennessee.edu> References: <5D2CC5653262C14188174E20065A4C4EC8D457@hscmbx6.uthsc.tennessee.edu> Message-ID: <3CC4394E-B87C-46A8-80C8-B002D2FF79C5@donders.ru.nl> Hi Xiaoxiao, It looks as if your mri (from which you obtained the vol-variable) and the data in the hs_file are not expressed in the same coordinate system. The geometrical objects that you need to combine in order to compute leadfields should be defined in the same coordinate system. My guess is that you will first need to apply ft_volumerealign to your mri_nom variable, to get it back into the Bti-based coordinate system. Best, Jan-Mathijs On Jul 21, 2012, at 1:28 AM, Bai, XiaoXiao wrote: > Dear Sir/Madam, > > I am a new fieldtrip user and want to apply the beamformer method in Fieldtrip for the MEG data from 4D/BTI system. Now I can get a individual subject MRI in analysis/spm format from CURRY software. > > This is my code for create volume head model for the beamformer, > > mri_nom = ft_read_mri(mrifile); > > cfg = []; > [segmentedmri] = ft_volumesegment(cfg, mri_nom); > > cfg = []; > cfg.method = 'singleshell'; > vol = ft_prepare_headmodel(cfg, segmentedmri); > > The vol and sensor locations were displayed and attached with here. > > How can I co-register the vol with sensors based on headshpe file (hs_file) for computing lead field for next step? > > Thanks a lot. > > Best regards, > > Xiaoxiao > Division of Clinical Neurosciences, Department of Pediatrics > University of Tennessee Health Science Center, College of Medicine & > Neuroscience Institute, LeBonheur Children's Hospital > 777 Washington Avenue, P335 > Memphis, TN 38105, USA > Phone: 901-287-4612 > Fax: 901-287-5325 > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Thu Aug 2 14:41:32 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Thu, 2 Aug 2012 14:41:32 +0200 Subject: [FieldTrip] prepare neighbors template file Message-ID: Dear Jörn, Thank you very much for fast answer. I have one more problem concerning this thing. When I do ft_timelockgrandaverage I got this warning: "discarding gradiometer information because it cannot be averaged". So, I tried different ways to calculate ft_megrealign. And I am not sure at which point I should do this function. Before trial averaging in single subject or after? Kind regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From lihqih at gmail.com Thu Aug 2 17:00:50 2012 From: lihqih at gmail.com (qi li) Date: Thu, 2 Aug 2012 11:00:50 -0400 Subject: [FieldTrip] mne source reconstruction In-Reply-To: <4570E04B-810A-4002-B286-561FDE5C0221@donders.ru.nl> References: <501A23BE.8040706@donders.ru.nl> <4570E04B-810A-4002-B286-561FDE5C0221@donders.ru.nl> Message-ID: Jan and Jörn: Thanks a lot for your answers! You did understand my questions pretty well. I actually tried to use ft_sourceinterpolate to find the voxel index without success. The error message is like 'Reference to non-existent field 'sphereradius'. Error in ft_sourceinterpolate (line 203) interpmat = interp_ungridded(functional.pos, warp_apply(anatomical.transform, [X(:) Y(:) Z(:)]), ...' On Thu, Aug 2, 2012 at 3:07 AM, jan-mathijs schoffelen wrote: > Hi Qi, > > To follow up on Jörn: > > If you reconstructed the source activations on a mesh that represents the > cortical surface, you first need to interpolate the functional data into a > 3D volume, if you want to use ft_sourceplot for visualization. For this you > need ft_sourceinterpolate. > As an alternative, to visualize the results, you can use ft_sourcemovie. > However, this function is under development so it may not give the most > visually appealing results. > > Best, > > JM > > > On Aug 2, 2012, at 8:52 AM, Jörn M. Horschig wrote: > > Dear Qi, > > Not quite sure if I understand your request correctly, but you can just > rehape: > source.avg.pow = reshape(source.avg.pow, anatomical.dim); > Then instead of being a vector, it will become a matrix. > > If that's not the answer you intended to get, please phrase your question > differently :) > > Best, > Jörn > > On 8/1/2012 7:33 PM, qi li wrote: > > Hi, > > Is there anyway to map the source space mesh back to voxel space so I > can use the ft_sourceplot. > > After MNE source reconstruction, I have > > source = > > time: [1x399 double] > pos: [8196x3 double] > inside: [1x8196 double] > outside: [1x0 double] > method: 'average' > avg: [1x1 struct] > cfg: [1x1 struct] > > I have 8196 mesh points but how to map it back to the anatomical voxel > space? Thanks! > > Qi > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > -- > Jörn M. Horschig > PhD Student > Donders Institute for Brain, Cognition and Behaviour > Centre for Cognitive Neuroimaging > Radboud University Nijmegen > Neuronal Oscillations Group > > P.O. Box 9101 > NL-6500 HB Nijmegen > The Netherlands > > Contact: > E-Mail: jm.horschig at donders.ru.nl > Tel: +31-(0)24-36-68493 > Web: http://www.ru.nl/donders > > Visiting address: > Trigon, room 2.30 > Kapittelweg 29 > NL-6525 EN Nijmegen > The Netherlands > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > Jan-Mathijs Schoffelen, MD PhD > > Donders Institute for Brain, Cognition and Behaviour, > Centre for Cognitive Neuroimaging, > Radboud University Nijmegen, The Netherlands > > Max Planck Institute for Psycholinguistics, > Nijmegen, The Netherlands > > J.Schoffelen at donders.ru.nl > Telephone: +31-24-3614793 > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From nenga at gmx.net Thu Aug 2 18:51:28 2012 From: nenga at gmx.net (Carina Oehrn) Date: Thu, 02 Aug 2012 18:51:28 +0200 Subject: [FieldTrip] granger causality: model order In-Reply-To: <17A8F077-820B-48A6-A388-2B2E0B7AB002@donders.ru.nl> References: <20120730162938.56630@gmx.net> <17A8F077-820B-48A6-A388-2B2E0B7AB002@donders.ru.nl> Message-ID: <20120802165128.149370@gmx.net> Hey Jan-Mathijs, thank you very much for your response and your advise on the length of time windows! TF-resolved Granger works fine with fieldtrip now. I still have my questions regarding the choice of model order and yes, sorry, my descriptions were not detailed enough. 1) what I meant was that I have problems using the aic_test function of the bsmart toolbox to calculate the most appropriate model order. When I am trying to use maximal model orders above 15 (I have 3000 data points and I tried various different time windows), I always get the error message: ??? Error using ==> aic_test at 47 Cannot compute AIC with 'opssaic'! But I can not find the reason. Others using this function seemed to have the same problem. However, as my data was sampled at a rate of 1000 Hz and as I do not downsample in my analysis, I would like to use orders above 15 (which to my knowledge most people use- who however mostly have smaller sampling rates) to include a time period longer than 15 ms. 2) What range of model orders were you talking about when stating: > For the model order: higher one. Some people seem to get nice results when taking a relatively high model order. Or does anybody else have some experience with higher model orders? How far up do you go to include a sufficient time interval but still avoid overparameterization? Thank you very much for your help! All the best, Carina -------- Original-Nachricht -------- > Datum: Thu, 2 Aug 2012 09:16:26 +0200 > Von: jan-mathijs schoffelen > An: FieldTrip discussion list > Betreff: Re: [FieldTrip] granger causality > Hi Carina, > > > 1.) I know that I can calculate the appropriate model order and window > length with the bsmart toolbox. But from the documentation I somehow do not > get the format of data I should use for this calculation. > > When I enter the time-frequency data (which I want to use to estimate > granger causality) (size: 2 x 2 x freq x time points), it doesn't work. > > The 'it doesn't work' statement is typically not enough for us to diagnose > where the problem lies (at least at which step it goes wrong). At the > moment I can only assume that you use an outdated version of FieldTrip, because > when I try to do TF-resolved Granger using AR-models it works. It is good > that you pasted the script below, but an error message may be informative > too. > > > Any advise? I have a sampling rate of 1000. What sort of time window and > maximal model order would make sense do you think? > > I think that the time window is rather short: I'd rather go for 0.4 or 0.5 > to begin with, but this is an empirical question. Also, shifting the > windows one time step at a time does not really make sense to me, 0.01 or even > 0.05 is much more memory friendly. For the model order: higher one. Some > people seem to get nice results when taking a relatively high model order. Of > course, for a high model order you also need longer time-windows. > > > 2)I would like to keep the time domain analyzing my data. What do you > think would be an appropriate sliding time window using the 'ft_mvaranalysis' > function? And what does it mean exactly? By choosing the model order, I am > already determining the maximal time lag between the two functions. Are > the values then estimated for the whole time window? > > No. The time window determines the length of the data segment which is > used to fit an AR-model (of order X) of the data at time point Y (as defined > in your toi) > > > 3)What do you think about statistics? Would it makes sense to use the > non-parametric cluster approach to shuffle within patients and do a group > analysis that way? > > This depends on your experimental question and you null-hypothesis. > > Best, > Jan-Mathijs > > > > > Thank you so much in advance! > > Best, > > Carina > > > > > > As a summary,I am doing following steps with field trip: > > > > > > cfg = []; > > cfg.dftfilter ='yes' > > prep_cond1{subj} = ft_preprocessing(cfg, data); > > > > > > cfg = []; > > cfg.order = 5; > > cfg.toolbox = 'bsmart'; > > cfg.t_ftimwin = 0.05 > > cfg.toi = -1:0.001:3.5; > > mdata_cond1{subj}= ft_mvaranalysis(cfg, prep_cond1{subj}); > > > > > > cfg = []; > > cfg.method = 'mvar'; > > cfg.foi = 4:100; > > cond1_freq{subj}= ft_freqanalysis_mvar(cfg,mdata_cond1{subj}); > > > > > > cfg = []; > > cfg.method = 'granger'; > > cond1_granger{subj} = ft_connectivityanalysis(cfg, cond1_freq{subj}); > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > Jan-Mathijs Schoffelen, MD PhD > > Donders Institute for Brain, Cognition and Behaviour, > Centre for Cognitive Neuroimaging, > Radboud University Nijmegen, The Netherlands > > Max Planck Institute for Psycholinguistics, > Nijmegen, The Netherlands > > J.Schoffelen at donders.ru.nl > Telephone: +31-24-3614793 > From fredericroux at hotmail.de Fri Aug 3 11:38:31 2012 From: fredericroux at hotmail.de (Frederic Roux) Date: Fri, 3 Aug 2012 11:38:31 +0200 Subject: [FieldTrip] PhD position Message-ID: PhD position Beginning in October 2012, the Brain Imaging Center, Department of Neurology, Goethe-University, offers a PhD position (3years, 65% of standard weekly hours) in the project: "Spectro-temporal dynamics of sensorimotor interactions underlying lateralization of speech production" funded by the German Research Council. The successful applicant will study speech production using magnetoencephalography, electro- corticography, and computational modelling. Our interdisciplinary Emmy Noether group focuses on the question how network dynamics contribute to the generation of complex behavior and offers - besides expertise in functional neuroimaging and Cognitive Neuroscience - a creative and cooperative work environment. We are seeking for an enthusiastic thesis student with interest in the Neurosciences who has advanced skills in Matlab programming and experience in signal processing and/or brain data analysis. He/she should have a background in Natural or Applied Science or Psychology. Electronic applications (no original documents) are requested until 2012/08/31 and should be addressed to: Dr. Christian Kell, Cognitive Neuroscience Group, Brain Imaging Center and Department of Neurology, Goethe University, Schleusenweg 2-16, 60528 Frankfurt, Germany c.kell at em.uni-frankfurt.de www.brainclocks.com Pay group E13 TV-GU. Teaching is not mandatory. Goethe University is an equal opportunity employer. Applications from female candidates are encouraged. Where qualifications are equal, preference will be given to people with disabilities. -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Fri Aug 3 15:36:53 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Fri, 3 Aug 2012 15:36:53 +0200 Subject: [FieldTrip] planar gradient literature Message-ID: Hi, I would like to ask for some literature concerning planar gradient computation. I am new in MEG field and I need some more information on theoretical background of such a calculation. The stuff available in "Event related averaging and planar gradient" tutorial is useful but not very detailed. So, could you please direct me to some reference. Thank you in advance! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From Lilla.Magyari at mpi.nl Fri Aug 3 16:10:31 2012 From: Lilla.Magyari at mpi.nl (Lilla.Magyari at mpi.nl) Date: Fri, 3 Aug 2012 16:10:31 +0200 (CEST) Subject: [FieldTrip] planar gradient literature In-Reply-To: References: Message-ID: <50094.131.174.45.70.1344003031.squirrel@131.174.45.70> Hi Nenad, I usually refer to this paper when I use the planar gradient computation: Bastiaansen, M. C. M., & Knösche, T. R. (2000). MEG tangential derivative mapping applied to Event-Related Desynchronization (ERD) research. Clinical Neurophysiology, 111, 1300-1305. Best, Lilla > Hi, > > I would like to ask for some literature concerning planar gradient > computation. I am new in MEG field and I need some more information on > theoretical background of such a calculation. The stuff available in > "Event > related averaging and planar gradient" tutorial is useful but not > very detailed. So, could you please direct me to some reference. > > Thank you in advance! > > Nenad > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From polomacnenad at gmail.com Fri Aug 3 18:45:48 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Fri, 3 Aug 2012 18:45:48 +0200 Subject: [FieldTrip] planar gradient literature Message-ID: Thank you Lilla! :) All the best! -------------- next part -------------- An HTML attachment was scrubbed... URL: From juan.garciaprieto at ctb.upm.es Sun Aug 5 20:41:35 2012 From: juan.garciaprieto at ctb.upm.es (Juan Garcia-Prieto) Date: Sun, 5 Aug 2012 20:41:35 +0200 Subject: [FieldTrip] anonymize registers Message-ID: Hi everybody, First time mailing this list. great pleasure. This is Juan, from madrid's MEG lab. I've coded a little matlab-FT script which anonymizes fif files (elekta neuromag). Brief story is that when sending a .fif file somewhere else for post-processing, it might be wise to eliminate any sensible information in terms subjects name, id, etc. because of data privacy concerns. When dealing with this issue, Elekta very gently supplied a script wrapper based on regular expressions, to be run under linux. So my effort will only be another way of solving the problem. The script uses "fiff_read_tag_info" "fiff_read_tag" "fiff_make_dir_tree" and "fiff_dir_tree_find" functions (part of FT) by Matti Hamalainen. My question, is: Will this function be useful to some others? Specially taking into account I have ONLY worked with neuromag .fif files. thank you for your work. I've learned a LOT, only reading FT scripts. Best regards, Juan, -------------- next part -------------- An HTML attachment was scrubbed... URL: From kashyapmanik at gmail.com Tue Aug 7 08:38:07 2012 From: kashyapmanik at gmail.com (Manik Gupta) Date: Tue, 7 Aug 2012 07:38:07 +0100 Subject: [FieldTrip] Help needed for reading a time series data Message-ID: Dear All I am new to fieldtrip and want to use to preprocess some air pollution data using fieldtrip. The data is sampled at a frequency of 1 sample/second and currently is in a text file format. Can someone please help me how to process the text file into fieldtrip raw data structure. Thanks a lot, Manik. -------------- next part -------------- An HTML attachment was scrubbed... URL: From stan.vanpelt at fcdonders.ru.nl Tue Aug 7 09:43:38 2012 From: stan.vanpelt at fcdonders.ru.nl (Stan van Pelt) Date: Tue, 7 Aug 2012 09:43:38 +0200 (CEST) Subject: [FieldTrip] Help needed for reading a time series data In-Reply-To: Message-ID: <29876546.58683.1344325418384.JavaMail.root@sculptor.zimbra.ru.nl> Dear Manik, Great that you're trying to apply FieldTrip to non-neuroscientific data. As you probably know, FT is developed for the analysis of electrophysiological data, but that does not mean it cannot be useful to you. W.r.t. your data format, i think it is easiest if you read in the data into Matlab using a function like readtxt.m, and than put these data into a FT-type structure, and run your actual analyses. At the minimum, this structure should look like this: data = label: {187x1 cell} % channel labels (e.g. 'air pollution site 1','air pollution site 2', etc) trial: {1x266 cell} % data (Nchans*Nsamples) for each trial (if you have only one long data set, it means you have only 1 trial; if you have multiple 'recording sessions' from the same site, you would have more) time: {1x266 cell} % time axis for each trial fsample: 300 % sampling frequency It would be great if you would be willing to share your results when they are published and add a FieldTrip reference (http://www.hindawi.com/journals/cin/2011/156869/). Good luck! Stan -- Stan van Pelt, PhD Ernst Strüngmann Institute (ESI) for Neuroscience in Cooperation with Max Planck Society Deutschordenstr. 46 60528 Frankfurt, Germany Website: www.esi-frankfurt.de E-mail: stan.vanpelt at esi-frankfurt.de Tel: +49 (0)69 96769 519 Fax: +49 (0)69 96769 555 ----- Oorspronkelijk bericht ----- > Van: "Manik Gupta" > Aan: fieldtrip at science.ru.nl > Verzonden: Dinsdag 7 augustus 2012 08:38:07 > Onderwerp: [FieldTrip] Help needed for reading a time series data > Dear All > I am new to fieldtrip and want to use to preprocess some air pollution > data using fieldtrip. > The data is sampled at a frequency of 1 sample/second and currently is > in a text file format. > Can someone please help me how to process the text file into fieldtrip > raw data structure. > Thanks a lot, > Manik. > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Stan van Pelt Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen Kapittelweg 29, 6525 EN Nijmegen, Netherlands E-mail: stan.vanpelt at donders.ru.nl Website: www.ru.nl/donders/ Phone: (+31) (0)24 36 10981 Fax: (+31) (0)24 36 10989 -------------- next part -------------- An HTML attachment was scrubbed... URL: From kashyapmanik at gmail.com Tue Aug 7 09:51:31 2012 From: kashyapmanik at gmail.com (Manik Gupta) Date: Tue, 7 Aug 2012 08:51:31 +0100 Subject: [FieldTrip] Help needed for reading a time series data In-Reply-To: <29876546.58683.1344325418384.JavaMail.root@sculptor.zimbra.ru.nl> References: <29876546.58683.1344325418384.JavaMail.root@sculptor.zimbra.ru.nl> Message-ID: Dear Stan, Thanks a lot for the reply. I actually want to do a transfer entropy analysis using Trentool and for that my data needs to be in the fieldtrip raw format. I figured out how to use the data now and will get back to you in case of more queries. Just a quick check, would you be able to provide any assistance on Trentool as well. and I would surely publish the results if they are encouraging enough!! :-) Thanks once again, Manik On Tue, Aug 7, 2012 at 8:43 AM, Stan van Pelt wrote: > Dear Manik, > > Great that you're trying to apply FieldTrip to non-neuroscientific data. > As you probably know, FT is developed for the analysis of > electrophysiological data, but that does not mean it cannot be useful to > you. > W.r.t. your data format, i think it is easiest if you read in the data > into Matlab using a function like readtxt.m, and than put these data into a > FT-type structure, and run your actual analyses. At the minimum, this > structure should look like this: > > data = > > label: {187x1 cell} % channel labels (e.g. 'air pollution site 1','air pollution site 2', etc) > trial: {1x266 cell} % data (Nchans*Nsamples) for each trial (if you have only one long data set, it means you have only 1 trial; if you have multiple 'recording sessions' from the same site, you would have more) > time: {1x266 cell} % time axis for each trial > fsample: 300 % sampling frequency > > It would be great if you would be willing to share your results when they > are published and add a FieldTrip reference ( > http://www.hindawi.com/journals/cin/2011/156869/). > > Good luck! > > Stan > > -- > Stan van Pelt, PhD > > Ernst Strüngmann Institute (ESI) > for Neuroscience in Cooperation with Max Planck Society > Deutschordenstr. 46 > 60528 Frankfurt, Germany > Website: www.esi-frankfurt.de > E-mail: stan.vanpelt at esi-frankfurt.de > Tel: +49 (0)69 96769 519 > Fax: +49 (0)69 96769 555 > > > ------------------------------ > > *Van: *"Manik Gupta" > *Aan: *fieldtrip at science.ru.nl > *Verzonden: *Dinsdag 7 augustus 2012 08:38:07 > *Onderwerp: *[FieldTrip] Help needed for reading a time series data > > > Dear All > > I am new to fieldtrip and want to use to preprocess some air pollution > data using fieldtrip. > The data is sampled at a frequency of 1 sample/second and currently is in > a text file format. > Can someone please help me how to process the text file into fieldtrip raw > data structure. > > Thanks a lot, > Manik. > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > -- > Stan van Pelt > > Donders Institute for Brain, Cognition and Behaviour, Radboud University > Nijmegen > Kapittelweg 29, 6525 EN Nijmegen, Netherlands > E-mail: stan.vanpelt at donders.ru.nl > Website: www.ru.nl/donders/ > Phone: (+31) (0)24 36 10981 > Fax: (+31) (0)24 36 10989 > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -------------- next part -------------- An HTML attachment was scrubbed... URL: From johanna.zumer at donders.ru.nl Tue Aug 7 14:12:20 2012 From: johanna.zumer at donders.ru.nl (Johanna Zumer) Date: Tue, 7 Aug 2012 14:12:20 +0200 Subject: [FieldTrip] mne source reconstruction In-Reply-To: References: <501A23BE.8040706@donders.ru.nl> <4570E04B-810A-4002-B286-561FDE5C0221@donders.ru.nl> Message-ID: Dear Qi, Are you using the most current FieldTrip version? That line is 217 in the current version. But cfg.sphereradius should be set a few lines previous to that line, so I'm not sure how it is a non-existent field. What are the cfg options you use to call ft_sourceinterpolate? Best, Johanna 2012/8/2 qi li > Jan and Jörn: > > Thanks a lot for your answers! You did understand my questions pretty > well. I actually tried to use ft_sourceinterpolate to find the voxel > index without success. The error message is like > > 'Reference to non-existent field 'sphereradius'. > > Error in ft_sourceinterpolate (line 203) > interpmat = interp_ungridded(functional.pos, > warp_apply(anatomical.transform, [X(:) Y(:) Z(:)]), ...' > > On Thu, Aug 2, 2012 at 3:07 AM, jan-mathijs schoffelen > wrote: > > Hi Qi, > > > > To follow up on Jörn: > > > > If you reconstructed the source activations on a mesh that represents the > > cortical surface, you first need to interpolate the functional data into > a > > 3D volume, if you want to use ft_sourceplot for visualization. For this > you > > need ft_sourceinterpolate. > > As an alternative, to visualize the results, you can use ft_sourcemovie. > > However, this function is under development so it may not give the most > > visually appealing results. > > > > Best, > > > > JM > > > > > > On Aug 2, 2012, at 8:52 AM, Jörn M. Horschig wrote: > > > > Dear Qi, > > > > Not quite sure if I understand your request correctly, but you can just > > rehape: > > source.avg.pow = reshape(source.avg.pow, anatomical.dim); > > Then instead of being a vector, it will become a matrix. > > > > If that's not the answer you intended to get, please phrase your question > > differently :) > > > > Best, > > Jörn > > > > On 8/1/2012 7:33 PM, qi li wrote: > > > > Hi, > > > > Is there anyway to map the source space mesh back to voxel space so I > > can use the ft_sourceplot. > > > > After MNE source reconstruction, I have > > > > source = > > > > time: [1x399 double] > > pos: [8196x3 double] > > inside: [1x8196 double] > > outside: [1x0 double] > > method: 'average' > > avg: [1x1 struct] > > cfg: [1x1 struct] > > > > I have 8196 mesh points but how to map it back to the anatomical voxel > > space? Thanks! > > > > Qi > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > > > > -- > > Jörn M. Horschig > > PhD Student > > Donders Institute for Brain, Cognition and Behaviour > > Centre for Cognitive Neuroimaging > > Radboud University Nijmegen > > Neuronal Oscillations Group > > > > P.O. Box 9101 > > NL-6500 HB Nijmegen > > The Netherlands > > > > Contact: > > E-Mail: jm.horschig at donders.ru.nl > > Tel: +31-(0)24-36-68493 > > Web: http://www.ru.nl/donders > > > > Visiting address: > > Trigon, room 2.30 > > Kapittelweg 29 > > NL-6525 EN Nijmegen > > The Netherlands > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > > Jan-Mathijs Schoffelen, MD PhD > > > > Donders Institute for Brain, Cognition and Behaviour, > > Centre for Cognitive Neuroimaging, > > Radboud University Nijmegen, The Netherlands > > > > Max Planck Institute for Psycholinguistics, > > Nijmegen, The Netherlands > > > > J.Schoffelen at donders.ru.nl > > Telephone: +31-24-3614793 > > > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -------------- next part -------------- An HTML attachment was scrubbed... URL: From cs at imm.dtu.dk Tue Aug 7 14:40:46 2012 From: cs at imm.dtu.dk (Carsten Stahlhut) Date: Tue, 7 Aug 2012 14:40:46 +0200 Subject: [FieldTrip] Release of the SmartphoneBrainScanner2 Message-ID: [Apologies for cross-posting] Dear Fieldtrip list We are very happy to announce the release of the SmartphoneBrainScanner2 platform: http://code.google.com/p/smartphonebrainscanner2/ SmartphoneBrainScanner2 is a framework for building cross-platform real-time EEG applications. Originally developed at the Technical University of Denmark for collecting and analyzing signals from Emotiv EPOC headset, its extensible architecture allows working with various EEG systems and multiple platforms. *Cross Platform* SmartphoneBrainScanner2 is written in Qt, a C++ framework offering the power of the native development and unified support for multiple platforms. Plus the UI can be created in QML, high-level declarative UI framework. SBS2 can be compiled for every platform supporting Qt 4, including Linux, OSX, Windows, Android, Maemo 5, MeeGo. Although not yet attempted, it should also work on iOS and BlackBerry OS. *Advanced EEG* SmartphoneBrainScanner2 contains state-of-the-art techniques for working with multi-channel EEG signal in real-time, most notably source reconstruction methods with online adaptation to the noise level. Current implemented source reconstruction approaches cover the minimum-norm and low resolution tomography (LORETA) methods formulated in a Bayesian framework using a expectation-maximization scheme for hyperparameter estimation. The SBS2 source reconstruction is realized using a pre-build forward model connecting the cortical surface with the electrodes at the scalp. The current forward model provided with the software is a 3-spheres model obtained from the Matlab toolbox SPM8 using coarse spatial resolution and with sensor positions in accordance with the Emotiv EPOC system Besides, source reconstruction methods, additional machine learning methods such as independent component analysis (ICA), common spatial patterns (CSP), and Bayesian classifiers are continuously added. *New Approach* Real-time EEG doesn't have to happen in the lab! Consumer-grade and inexpensive research neuroheadsets allow for portability, delivering high-quality EEG signal. SmartphoneBrainScanner2 apps can be developed just like any other apps, featuring reach interface, connectivity, etc. Go, create! *How to cite SBS2* Please acknowledge the work of the SmartphoneBrainScanner2 by citing (Stopczynski et al, 2011). See also additional project related references at the homepage. A. Stopczynski, J. E. Larsen, C. Stahlhut, M. K. Petersen, & L. K. Hansen (2011), *A smartphone interface for a wireless EEG headset with real-time 3D reconstruction*, Affective Computing and Intelligent Interaction (ACII 2011), Lecture Notes in Computer Science (LNCS) 6357, Springer-Verlag Berlin Heidelberg, pp.317-318. *How to import data to Matlab* Demo scripts of how to import data to Matlab and to convert data into EEGLAB, FieldTrip, or SPM8 format are provided. Further description can be found at the Data Handling page: http://code.google.com/p/smartphonebrainscanner2/wiki/DataHandling?ts=1344340750&updated=DataHandling *Developer team* - Arkadiusz Stopczynski, DTU Informatics - Carsten Stahlhut, DTU Informatics - Michael Kai Petersen, DTU Informatics - Jakob Eg Larsen, DTU Informatics - Lars Kai Hansen, DTU Informatics *Acknowledgement* This work is supported in part by - Danish Lundbeck Foundation through Center for Integrated Molecular Brain Imaging (CIMBI ) - Danish Lundbeck Foundation through the project Real-time brain imaging by EEG - H C Ørsted Foundation - Nokia Best, Carsten -- Carsten Stahlhut Section for Cognitive Systems Department of Informatics and Mathematical Modelling Richard Petersens Plads, Building 321 Technical University of Denmark DK-2800 Kongens Lyngby, Denmark -------------- next part -------------- An HTML attachment was scrubbed... URL: From stan.vanpelt at fcdonders.ru.nl Wed Aug 8 16:14:38 2012 From: stan.vanpelt at fcdonders.ru.nl (Stan van Pelt) Date: Wed, 8 Aug 2012 16:14:38 +0200 (CEST) Subject: [FieldTrip] Fwd: source localization - MNI coordinates In-Reply-To: <1193658750.1278539.1344433666547.JavaMail.root@indus.zimbra.ru.nl> Message-ID: <1326953416.73376.1344435278179.JavaMail.root@sculptor.zimbra.ru.nl> Hi Alina, I forward your question to the FT discussion list, because I do not have experience with interpreting the source localization coordinate values. e.g., I do not know if they are actually in MNI space. ----- Doorgestuurd bericht ----- > Van: "A.S. Peter (Alina)" > Hi Stan, > > by visual inspection, I always thought that my source localization was > quite ok, but now I actually looked with MRICron where the spm > coordinates (=MNI coordinates, right?) are that fieldtrip outputs, and > some are near occipital regions but outside the brain... I am puzzled. > I tried linear ft_volumenormalize and also nonlinear, the MNI > coordinates are somewhat different but the problem remains. > > I then went back to look at SAM which we did first and it seems there > the sources were localized somewhat deeper/nicer than with LCMV > anyhow, which also tells me it's probably not a fiducial problem. I > think it may be partly my code (i.e. what I basically took from Avgis > and looks like yours) but also partly conversion to MNI causing > problems. Did you ever try to see the MNI coordinates for your data? > > Best, > > Alina -- Stan van Pelt Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen Kapittelweg 29, 6525 EN Nijmegen, Netherlands E-mail: stan.vanpelt at donders.ru.nl Website: www.ru.nl/donders/ Phone: (+31) (0)24 36 10981 Fax: (+31) (0)24 36 10989 From polomacnenad at gmail.com Thu Aug 9 11:56:13 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Thu, 9 Aug 2012 11:56:13 +0200 Subject: [FieldTrip] ordinal numbers of bad trials Message-ID: Hi, I would like to know is there an option to save ordinal number of bad trials after the artifact rejection step. I noticed that in all artifact rejection methods(ft_rejectvisual or ft_artifact_zvalue) I end up with structure "variable_name.cfg.artifact"which consist time points of the bad segments but I cannot find ordinal numbers of it. Kind regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From cornabel at googlemail.com Thu Aug 9 13:54:59 2012 From: cornabel at googlemail.com (cornelius abel) Date: Thu, 09 Aug 2012 13:54:59 +0200 Subject: [FieldTrip] ordinal numbers of bad trials In-Reply-To: References: Message-ID: <5023A513.7090208@googlemail.com> Hi Nenad, for the reject visual function i have a work around to this problem. First you have to add a column to the data.trialinfo table with a unique trial ID (in my case column 6). Then you can compare the output trialinfo of the rejectvisual function (dummy.trialinfo) to the original trialinfo and identify the rejected trials. In VisBadTrials you then have the unique trial ID of the rejected trials. Have a look at the code snip: ... dummy = ft_rejectvisual(cfg,data); VisBadTrials=data.trialinfo((find(~ismember(data.trialinfo(:,6),dummy.trialinfo(:,6)))),6)'; VisBadTrials=unique(VisBadTrials); Cornelius Am 09.08.2012 11:56, schrieb Nenad Polomac: > Hi, > > I would like to know is there an option to save ordinal number of bad > trials after the artifact rejection step. I noticed that in all > artifact rejection methods(ft_rejectvisual or ft_artifact_zvalue) I > end up with structure "variable_name.cfg.artifact"which consist time > points of the bad segments but I cannot find ordinal numbers of it. > > Kind regards! > > Nenad > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From Markus.Butz at uni-duesseldorf.de Thu Aug 9 17:24:48 2012 From: Markus.Butz at uni-duesseldorf.de (Markus Butz) Date: Thu, 09 Aug 2012 16:24:48 +0100 Subject: [FieldTrip] ft_rejectvisual: channel display Message-ID: <7570de0d8d01.5023e450@uni-duesseldorf.de> Dear list After defining 1 sec trials from continuous CTF data I want to use ft_rejectvisual using the following bit of code: cfg.method = 'trial'; chansel = ft_channelselection({'M*','EOG*','-B*','-P*','-R*','-Q*'}, data.label); cfg.channel = chansel; cfg.keepchannel = 'no'; data = ft_rejectvisual(cfg,data); What I don't understand is, why all channels are displayed in the layout, while only the selected ones show data and are saved in data.label (after going through >1 trials). However, when recalling "data = ft_rejectvisual(cfg,data);" only the selected ones are displayed in the layout with data. Does anyone has an explanation/fix for this, ie a solution that solely selected channels are displayed with the first call of ft_rejectvisual? Cheers Markus -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Fri Aug 10 14:43:23 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Fri, 10 Aug 2012 14:43:23 +0200 Subject: [FieldTrip] ordinal numbers of bad trials Message-ID: Dear Cornelius, Thank you very much for this cool solution! Kind regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From sauer.mpih at googlemail.com Fri Aug 10 15:49:47 2012 From: sauer.mpih at googlemail.com (Andreas Sauer) Date: Fri, 10 Aug 2012 15:49:47 +0200 Subject: [FieldTrip] ft_qualitycheck Message-ID: Dear all, I tried to check my MEG (CTF) datasets with ft_qualitycheck. Unfortunately it didn't work. An error occurs in the boxplot.m function in line 2020, where the data should be copied into the dataset. But for me, there is no dataset(). My question is whether this is due to my dataset-information or due to an error in the script? An how to solve this issue? Any help is highly appreciated! Thanks! Best regards, Andreas -- Andreas Sauer Max Planck Institute for Brain Research Department of Neurophysiology Deutschordenstraße 46 60528 Frankfurt am Main Germany T: +49 69 96769 278 F: +49 69 96769 327 Email: sauer.mpih at gmail.com www.brain.mpg.de -------------- next part -------------- An HTML attachment was scrubbed... URL: From a.stolk at fcdonders.ru.nl Fri Aug 10 17:17:06 2012 From: a.stolk at fcdonders.ru.nl (Stolk, A.) Date: Fri, 10 Aug 2012 17:17:06 +0200 (CEST) Subject: [FieldTrip] ft_qualitycheck In-Reply-To: Message-ID: <1868283926.90397.1344611826092.JavaMail.root@sculptor.zimbra.ru.nl> Hi Andreas, The boxplot function is a matlab statistics toolbox function. I do not see why it throws an error here. However, I did find a problem  related to the dataset history file (.hist) which contains details of the recording itself. I will fix this issue  right away. If you are still getting the same error  with tomorrow's version of FT, please give me shout. And preferably  provide some more details about the error, and how you are calling the function. best, Arjen ----- Oorspronkelijk bericht ----- > Van: "Andreas Sauer" > Aan: fieldtrip at science.ru.nl > Verzonden: Vrijdag 10 augustus 2012 15:49:47 > Onderwerp: [FieldTrip] ft_qualitycheck > Dear all, > I tried to check my MEG (CTF) datasets with ft_qualitycheck. > Unfortunately it didn't work. An error occurs in the boxplot.m > function in line 2020, where the data should be copied into the > dataset. But for me, there is no dataset(). > My question is whether this is due to my dataset-information or due to > an error in the script? An how to solve this issue? > Any help is highly appreciated! > Thanks! > Best regards, > Andreas > -- > Andreas Sauer > Max Planck Institute for Brain Research > Department of Neurophysiology > Deutschordenstraße 46 > 60528 Frankfurt am Main > Germany > T: +49 69 96769 278 > F: +49 69 96769 327 > Email: sauer.mpih at gmail.com > www.brain.mpg.de > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From sauer.mpih at googlemail.com Mon Aug 13 11:39:23 2012 From: sauer.mpih at googlemail.com (Andreas Sauer) Date: Mon, 13 Aug 2012 11:39:23 +0200 Subject: [FieldTrip] ft_qualitycheck In-Reply-To: <1868283926.90397.1344611826092.JavaMail.root@sculptor.zimbra.ru.nl> References: <1868283926.90397.1344611826092.JavaMail.root@sculptor.zimbra.ru.nl> Message-ID: Dear Arjen, thanks for your replay! I just tried with the current version of fieldtrip (20120812) but I still get this error message: ??? Undefined function or variable 'dataset'. Error in ==> boxplot>computeBoxLocation at 2020 boxValDs = dataset(); Error in ==> boxplot at 312 [boxVal,boxValPlot,boxIdx] = computeBoxLocation(xSorted,... Error in ==> ft_qualitycheck>draw_figure at 547 boxplot(h.HmotionAxes, hmotions, 'orientation', 'horizontal', 'notch', 'on'); Error in ==> ft_qualitycheck at 296 draw_figure(info, temp_timelock, temp_freq, temp_summary, temp_headpos, toi); I call the function ft_qualitycheck as described in the tutorial. cfg.dataset = subject_xy.ds; ft_qualitycheck(cfg). It runs the analysis, it writes the .mat-file but then it throws the error message. So to me, it seems that there is some information (the variable 'dataset') missing. But I don't know whether this is due to our recording or an error in one of the functions... Best, Andreas 2012/8/10 Stolk, A. > Hi Andreas, > > > > The boxplot function is a matlab statistics toolbox function. I do not see > why it throws an error here. However, I did find a problem related to the > dataset history file (.hist) which contains details of the recording > itself. I will fix this issue right away. If you are still getting the same > error with tomorrow's version of FT, please give me shout. And > preferably provide some more details about the error, and how you are > calling the function. > > > > best, > > Arjen > > > > > ------------------------------ > > *Van: *"Andreas Sauer" > *Aan: *fieldtrip at science.ru.nl > *Verzonden: *Vrijdag 10 augustus 2012 15:49:47 > *Onderwerp: *[FieldTrip] ft_qualitycheck > > > Dear all, > > I tried to check my MEG (CTF) datasets with ft_qualitycheck. Unfortunately > it didn't work. An error occurs in the boxplot.m function in line 2020, > where the data should be copied into the dataset. But for me, there is no > dataset(). > > My question is whether this is due to my dataset-information or due to an > error in the script? An how to solve this issue? > > Any help is highly appreciated! > > Thanks! > > Best regards, > > Andreas > > > -- > Andreas Sauer > Max Planck Institute for Brain Research > Department of Neurophysiology > Deutschordenstraße 46 > 60528 Frankfurt am Main > Germany > > T: +49 69 96769 278 > F: +49 69 96769 327 > Email: sauer.mpih at gmail.com > www.brain.mpg.de > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- Dipl.-Psych. Andreas Sauer Max Planck Institute for Brain Research Deutschordenstraße 46 60528 Frankfurt am Main Germany T: +49 69 96769 278 F: +49 69 96769 327 Email: sauer.mpih at gmail.com www.brain.mpg.de -------------- next part -------------- An HTML attachment was scrubbed... URL: From a.stolk at fcdonders.ru.nl Mon Aug 13 13:19:11 2012 From: a.stolk at fcdonders.ru.nl (Stolk, A.) Date: Mon, 13 Aug 2012 13:19:11 +0200 (CEST) Subject: [FieldTrip] ft_qualitycheck In-Reply-To: Message-ID: <1337301516.103006.1344856751003.JavaMail.root@sculptor.zimbra.ru.nl> Hi Andreas, Thanks for testing with the newest version. My suspicion that your bug pertains to the matlab function 'boxplot' has grown. Boxplot calls another function, 'dataset', that does exist in my version of matlab (2010b). To be more precise, it is located in the /toolbox/shared/statslib/@dataset/ directory. In my version of boxplot the line 'boxValDs = dataset();' is at line 2021 and not at line 2020. Please therefore check if your boxplot version, or your matlab version in general, is not corrupted by performing the following command: boxplot(1). Hope this helps, Arjen ----- Oorspronkelijk bericht ----- > Van: "Andreas Sauer" > Aan: "FieldTrip discussion list" > Verzonden: Maandag 13 augustus 2012 11:39:23 > Onderwerp: Re: [FieldTrip] ft_qualitycheck > Dear Arjen, > thanks for your replay! > I just tried with the current version of fieldtrip (20120812) but I > still get this error message: > ??? Undefined function or variable 'dataset'. > Error in ==> boxplot>computeBoxLocation at 2020 > boxValDs = dataset(); > Error in ==> boxplot at 312 > [boxVal,boxValPlot,boxIdx] = computeBoxLocation(xSorted,... > Error in ==> ft_qualitycheck>draw_figure at 547 >   boxplot(h.HmotionAxes, hmotions, 'orientation', 'horizontal', > 'notch', 'on'); > Error in ==> ft_qualitycheck at 296 >       draw_figure(info, temp_timelock, temp_freq, temp_summary, > temp_headpos, toi); > I call the function ft_qualitycheck as described in the tutorial. > cfg.dataset = subject_xy.ds; ft_qualitycheck(cfg). It runs the > analysis, it writes the .mat-file but then it throws the error > message. So to me, it seems that there is some information (the > variable 'dataset') missing. But I don't know whether this is due to > our recording or an error in one of the functions... > Best, > Andreas > 2012/8/10 Stolk, A. < a.stolk at fcdonders.ru.nl > > > Hi Andreas, > >   > > The boxplot function is a matlab statistics toolbox function. I do > > not > > see why it throws an error here. However, I did find a > > problem related > > to the dataset history file (.hist) which contains details of the > > recording itself. I will fix this issue right away. If you are still > > getting the same error with tomorrow's version of FT, please give me > > shout. And preferably provide some more details about the error, and > > how you are calling the function. > >   > > best, > > Arjen > >   > > > Van: "Andreas Sauer" < sauer.mpih at googlemail.com > > > > Aan: fieldtrip at science.ru.nl > > > Verzonden: Vrijdag 10 augustus 2012 15:49:47 > > > Onderwerp: [FieldTrip] ft_qualitycheck > > > Dear all, > > > I tried to check my MEG (CTF) datasets with ft_qualitycheck. > > > Unfortunately it didn't work. An error occurs in the boxplot.m > > > function in line 2020, where the data should be copied into the > > > dataset. But for me, there is no dataset(). > > > My question is whether this is due to my dataset-information or > > > due > > > to > > > an error in the script? An how to solve this issue? > > > Any help is highly appreciated! > > > Thanks! > > > Best regards, > > > Andreas > > > -- > > > Andreas Sauer > > > Max Planck Institute for Brain Research > > > Department of Neurophysiology > > > Deutschordenstraße 46 > > > 60528 Frankfurt am Main > > > Germany > > > T: +49 69 96769 278 > > > F: +49 69 96769 327 > > > Email: sauer.mpih at gmail.com > > > www.brain.mpg.de > > > _______________________________________________ > > > fieldtrip mailing list > > > fieldtrip at donders.ru.nl > > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- > Dipl.-Psych. Andreas Sauer > Max Planck Institute for Brain Research > Deutschordenstraße 46 > 60528 Frankfurt am Main > Germany > T: +49 69 96769 278 > F: +49 69 96769 327 > Email: sauer.mpih at gmail.com > www.brain.mpg.de > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From eelke.spaak at donders.ru.nl Mon Aug 13 13:30:36 2012 From: eelke.spaak at donders.ru.nl (Eelke Spaak) Date: Mon, 13 Aug 2012 13:30:36 +0200 Subject: [FieldTrip] ft_qualitycheck In-Reply-To: <1337301516.103006.1344856751003.JavaMail.root@sculptor.zimbra.ru.nl> References: <1337301516.103006.1344856751003.JavaMail.root@sculptor.zimbra.ru.nl> Message-ID: Hi Andreas, In addition to Arjen's comments, it might also be wise to execute 'which boxplot' on the Matlab command prompt, to see whether you have a version of boxplot on your path that is shadowing the stats toolbox version. Best, Eelke On 13 August 2012 13:19, Stolk, A. wrote: > Hi Andreas, > > > > Thanks for testing with the newest version. My suspicion that your bug > pertains to the matlab function 'boxplot' has grown. Boxplot calls another > function, 'dataset', that does exist in my version of matlab (2010b). To be > more precise, it is located in the /toolbox/shared/statslib/@dataset/ > directory. > > > > In my version of boxplot the line 'boxValDs = dataset();' is at line 2021 > and not at line 2020. Please therefore check if your boxplot version, or > your matlab version in general, is not corrupted by performing the following > command: boxplot(1). > > > > Hope this helps, > > Arjen > > ________________________________ > > Van: "Andreas Sauer" > Aan: "FieldTrip discussion list" > Verzonden: Maandag 13 augustus 2012 11:39:23 > Onderwerp: Re: [FieldTrip] ft_qualitycheck > > > Dear Arjen, > > thanks for your replay! > > I just tried with the current version of fieldtrip (20120812) but I still > get this error message: > > > ??? Undefined function or variable 'dataset'. > > Error in ==> boxplot>computeBoxLocation at 2020 > boxValDs = dataset(); > > Error in ==> boxplot at 312 > [boxVal,boxValPlot,boxIdx] = computeBoxLocation(xSorted,... > > Error in ==> ft_qualitycheck>draw_figure at 547 > boxplot(h.HmotionAxes, hmotions, 'orientation', 'horizontal', 'notch', > 'on'); > > Error in ==> ft_qualitycheck at 296 > draw_figure(info, temp_timelock, temp_freq, temp_summary, > temp_headpos, toi); > > > I call the function ft_qualitycheck as described in the tutorial. > cfg.dataset = subject_xy.ds; ft_qualitycheck(cfg). It runs the analysis, it > writes the .mat-file but then it throws the error message. So to me, it > seems that there is some information (the variable 'dataset') missing. But I > don't know whether this is due to our recording or an error in one of the > functions... > > Best, > > Andreas > > > > > 2012/8/10 Stolk, A. >> >> Hi Andreas, >> >> >> >> The boxplot function is a matlab statistics toolbox function. I do not see >> why it throws an error here. However, I did find a problem related to the >> dataset history file (.hist) which contains details of the recording itself. >> I will fix this issue right away. If you are still getting the same error >> with tomorrow's version of FT, please give me shout. And preferably provide >> some more details about the error, and how you are calling the function. >> >> >> >> best, >> >> Arjen >> >> >> >> >> ________________________________ >> >> Van: "Andreas Sauer" >> Aan: fieldtrip at science.ru.nl >> Verzonden: Vrijdag 10 augustus 2012 15:49:47 >> Onderwerp: [FieldTrip] ft_qualitycheck >> >> >> Dear all, >> >> I tried to check my MEG (CTF) datasets with ft_qualitycheck. Unfortunately >> it didn't work. An error occurs in the boxplot.m function in line 2020, >> where the data should be copied into the dataset. But for me, there is no >> dataset(). >> >> My question is whether this is due to my dataset-information or due to an >> error in the script? An how to solve this issue? >> >> Any help is highly appreciated! >> >> Thanks! >> >> Best regards, >> >> Andreas >> >> >> -- >> Andreas Sauer >> Max Planck Institute for Brain Research >> Department of Neurophysiology >> Deutschordenstraße 46 >> 60528 Frankfurt am Main >> Germany >> >> T: +49 69 96769 278 >> F: +49 69 96769 327 >> Email: sauer.mpih at gmail.com >> www.brain.mpg.de >> >> >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip >> >> >> >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > -- > Dipl.-Psych. Andreas Sauer > Max Planck Institute for Brain Research > Deutschordenstraße 46 > 60528 Frankfurt am Main > Germany > > T: +49 69 96769 278 > F: +49 69 96769 327 > Email: sauer.mpih at gmail.com > www.brain.mpg.de > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From maximilien.chaumon at gmail.com Mon Aug 13 14:16:56 2012 From: maximilien.chaumon at gmail.com (Maximilien Chaumon) Date: Mon, 13 Aug 2012 14:16:56 +0200 Subject: [FieldTrip] [Eeglablist] nested hypothesis testing to decide whether to use one or two dipoles to fit a component In-Reply-To: References: Message-ID: Hello eeglab & Fieldtrip, I'm trying to find out if it would be possible to use a nested hypothesis testing approach to decide whether to use a one or two dipole model while estimating components' dipole locations. The rationale I would like to follow is this: with two dipoles, we will always obtain a better fit than with one dipole, but the decrease in sum of squared errors (SSE) should follow a F distribution with k (= Nparameters_2dipoles - Nparameters_1dipole) degrees of freedom. If the decrease in SSE is greater than what would be expected under this F distribution, then we decide that 2 dipoles provide a sufficiently better fit and decide using them. I asked this question to eeglablist and Scott pointed out that it is difficult/impossible(?) to determine if the second dipole fits actual interesting data or just noise introduced by the imperfect head model. Christian then said it'd be worth a shot, and I agree, so here I am again with two questions, or two confirmations: 1) *How many parameters are estimated in ft_dipolefitting.m ?* specially in the case of 2 dipoles. If I count correctly, we estimate 6 parameters for one dipole, and, depending on whether the orientation has to be the same in the 2 dipoles, one (amplitude) or three (amplitude and orientation) more. 2) *Can I assimilate the relative residual variance to a SSE?* the function rv.m does this: rv = sum((d1-d2).^2) ./ sum(d1.^2); So that seems to be a sum of squared errors divided by the variance of the original data. So if I multiply the rv by the sum squared component map, I should get it, right? Thanks a lot! Max 2012/8/11 Christian Kothe > I can only speak from my armchair here, but it sounds like it should be > worth a try - even if you don't get the # of parameters exactly right it > will probably give you at least some level of complexity control in > whatever the range of validity is. If it works, it may inspire follow-up > work (e.g., Bayesian model selection or likelihood ratio tests). > > The number of parameters for a 2-dipole model seems to be 3 (xyz) + 4 (2x > the orientation parameters). Not sure about the momentum, though - you > might look up the place where the actual function minimization is being > performed in dipfit (fminunc call?) and see whether these are being > optimized together with the others. > > Christian > > On Aug 10, 2012, at 3:29 PM, Scott Makeig wrote: > > MAX - Unfortunately, in general using two dipoles rather than one will > ~always improve the fit. Even if the source is a pure single dipole, a > second dipole can be used to correct for noise or errors in the forward > head model. This is less always the case for the constrained > spatially-symmetric dipole pairs allowed by dipfit(). However, we have not > thought of an optimal way to decide between using one or (occasionally) two > dipoles to fit e.g. maps of ICA brain sources. The goal would be to decide > whether the two-dipole version is fitting noise/forward model error vs > actual bilateral source generation... > > Scott Makeig > > On Thu, Aug 9, 2012 at 1:54 AM, Maximilien Chaumon < > maximilien.chaumon at gmail.com> wrote: > >> Hello all, >> >> When fitting dipoles to components, we are all sooner or later puzzled by >> the question whether to use one or two symmetrical dipoles. >> >> Would it be correct to put the problems in terms of a nested hypothesis >> testing? >> >> We are fitting a scalp map with one or two parameters and get a residual >> variance after the fit. >> Could we not use this residual variance as a measure of the SSE and >> compute a F statistic to decide whether to use the more complex (with two >> dipoles) or simpler (with one dipole) of two nested models? >> If yes, then how would we decide on the number of degrees of freedom? How >> many free parameters do we have in each case? x,y,z,and two orientations >> per dipole? how does the imposed symmetry affect that number? Could we >> really map residual variance to SSE? How many "observations" do we have in >> that case (see formula below)? >> >> I found this formula, for F: >> F = (SSEF-SSER)/ (kF-kR) / ((1-SSEF)/(N-kF-1)) >> where >> SSE is sum of squared errors, >> k is numbers of parameters, >> N is number of observations (? what in our case?) >> F and R indices for full and reduced model respectively (in our case two >> and one dipole). >> >> >> Thanks a lot for any comment! >> Best, >> Max >> >> dipfit >> >> >> >> >> >> _______________________________________________ >> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html >> To unsubscribe, send an empty email to >> eeglablist-unsubscribe at sccn.ucsd.edu >> For digest mode, send an email with the subject "set digest mime" to >> eeglablist-request at sccn.ucsd.edu >> > > > > -- > Scott Makeig, Research Scientist and Director, Swartz Center for > Computational Neuroscience, Institute for Neural Computation; Prof. of > Neurosciences (Adj.), University of California San Diego, La Jolla CA > 92093-0559, http://sccn.ucsd.edu/~scott > > _______________________________________________ > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html > To unsubscribe, send an empty email to > eeglablist-unsubscribe at sccn.ucsd.edu > For digest mode, send an email with the subject "set digest mime" to > eeglablist-request at sccn.ucsd.edu > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From sauer.mpih at googlemail.com Mon Aug 13 14:59:09 2012 From: sauer.mpih at googlemail.com (Andreas Sauer) Date: Mon, 13 Aug 2012 14:59:09 +0200 Subject: [FieldTrip] ft_qualitycheck In-Reply-To: References: <1337301516.103006.1344856751003.JavaMail.root@sculptor.zimbra.ru.nl> Message-ID: Dear Eelke, dear Arjen, thanks for helping! Here, I analyze data on a cluster with Matlab 2008b. Although I found a folder "@dataset" in /toolbox/shared/statslib/, unfortunately there is no such function in it. Since I found the function in my local Matlab, which is version 2010a, I guess that this is a feature of newer Matlab versions. I downloaded one dataset to my local machine and with Matlab 2010a it worked... I am not sure whether it might be good to note that in the tutorial... Best, Andreas 2012/8/13 Eelke Spaak > Hi Andreas, > > In addition to Arjen's comments, it might also be wise to execute > 'which boxplot' on the Matlab command prompt, to see whether you have > a version of boxplot on your path that is shadowing the stats toolbox > version. > > Best, > Eelke > > On 13 August 2012 13:19, Stolk, A. wrote: > > Hi Andreas, > > > > > > > > Thanks for testing with the newest version. My suspicion that your bug > > pertains to the matlab function 'boxplot' has grown. Boxplot calls > another > > function, 'dataset', that does exist in my version of matlab (2010b). To > be > > more precise, it is located in the /toolbox/shared/statslib/@dataset/ > > directory. > > > > > > > > In my version of boxplot the line 'boxValDs = dataset();' is at line 2021 > > and not at line 2020. Please therefore check if your boxplot version, or > > your matlab version in general, is not corrupted by performing the > following > > command: boxplot(1). > > > > > > > > Hope this helps, > > > > Arjen > > > > ________________________________ > > > > Van: "Andreas Sauer" > > Aan: "FieldTrip discussion list" > > Verzonden: Maandag 13 augustus 2012 11:39:23 > > Onderwerp: Re: [FieldTrip] ft_qualitycheck > > > > > > Dear Arjen, > > > > thanks for your replay! > > > > I just tried with the current version of fieldtrip (20120812) but I still > > get this error message: > > > > > > ??? Undefined function or variable 'dataset'. > > > > Error in ==> boxplot>computeBoxLocation at 2020 > > boxValDs = dataset(); > > > > Error in ==> boxplot at 312 > > [boxVal,boxValPlot,boxIdx] = computeBoxLocation(xSorted,... > > > > Error in ==> ft_qualitycheck>draw_figure at 547 > > boxplot(h.HmotionAxes, hmotions, 'orientation', 'horizontal', 'notch', > > 'on'); > > > > Error in ==> ft_qualitycheck at 296 > > draw_figure(info, temp_timelock, temp_freq, temp_summary, > > temp_headpos, toi); > > > > > > I call the function ft_qualitycheck as described in the tutorial. > > cfg.dataset = subject_xy.ds; ft_qualitycheck(cfg). It runs the analysis, > it > > writes the .mat-file but then it throws the error message. So to me, it > > seems that there is some information (the variable 'dataset') missing. > But I > > don't know whether this is due to our recording or an error in one of the > > functions... > > > > Best, > > > > Andreas > > > > > > > > > > 2012/8/10 Stolk, A. > >> > >> Hi Andreas, > >> > >> > >> > >> The boxplot function is a matlab statistics toolbox function. I do not > see > >> why it throws an error here. However, I did find a problem related to > the > >> dataset history file (.hist) which contains details of the recording > itself. > >> I will fix this issue right away. If you are still getting the same > error > >> with tomorrow's version of FT, please give me shout. And preferably > provide > >> some more details about the error, and how you are calling the function. > >> > >> > >> > >> best, > >> > >> Arjen > >> > >> > >> > >> > >> ________________________________ > >> > >> Van: "Andreas Sauer" > >> Aan: fieldtrip at science.ru.nl > >> Verzonden: Vrijdag 10 augustus 2012 15:49:47 > >> Onderwerp: [FieldTrip] ft_qualitycheck > >> > >> > >> Dear all, > >> > >> I tried to check my MEG (CTF) datasets with ft_qualitycheck. > Unfortunately > >> it didn't work. An error occurs in the boxplot.m function in line 2020, > >> where the data should be copied into the dataset. But for me, there is > no > >> dataset(). > >> > >> My question is whether this is due to my dataset-information or due to > an > >> error in the script? An how to solve this issue? > >> > >> Any help is highly appreciated! > >> > >> Thanks! > >> > >> Best regards, > >> > >> Andreas > >> > >> > >> -- > >> Andreas Sauer > >> Max Planck Institute for Brain Research > >> Department of Neurophysiology > >> Deutschordenstraße 46 > >> 60528 Frankfurt am Main > >> Germany > >> > >> T: +49 69 96769 278 > >> F: +49 69 96769 327 > >> Email: sauer.mpih at gmail.com > >> www.brain.mpg.de > >> > >> > >> _______________________________________________ > >> fieldtrip mailing list > >> fieldtrip at donders.ru.nl > >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > >> > >> > >> > >> _______________________________________________ > >> fieldtrip mailing list > >> fieldtrip at donders.ru.nl > >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > > > > > > -- > > Dipl.-Psych. Andreas Sauer > > Max Planck Institute for Brain Research > > Deutschordenstraße 46 > > 60528 Frankfurt am Main > > Germany > > > > T: +49 69 96769 278 > > F: +49 69 96769 327 > > Email: sauer.mpih at gmail.com > > www.brain.mpg.de > > > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- Dipl.-Psych. Andreas Sauer Max Planck Institute for Brain Research Deutschordenstraße 46 60528 Frankfurt am Main Germany T: +49 69 96769 278 F: +49 69 96769 327 Email: sauer.mpih at gmail.com www.brain.mpg.de -------------- next part -------------- An HTML attachment was scrubbed... URL: From akiko.ikkai at gmail.com Tue Aug 14 22:19:13 2012 From: akiko.ikkai at gmail.com (Akiko Ikkai) Date: Tue, 14 Aug 2012 16:19:13 -0400 Subject: [FieldTrip] actvsblT and average over time and/or frequency Message-ID: Hi all, I'm testing for the power difference between baseline (fixation) and memory delay periods, following the examples in "Permutation Tests > Within trial experiments" tutorial. When I set cfg.statistic = 'actvsblT'; and cfg.avgovertime = 'yes'; cfg.avgoverfreq = 'yes'; then run ft_freqstatistics, I get an error in statfun_actvsblT.m, "Inappropriate dimord for the statistics function STATFUN_ACTVSBLT.". This is because with avgovertime and/or avgoverfreq set to 'yes', cfg.dimord for data input to statfun_actvsblT is simply "chan," whereas statfun_actvsblT requires "chan_freq_time." When I comment out cfg.avgovertime AND freq, permutation runs fine. This is particular to statfun_actvsblT; other functions, such as statfun_depsamplesT, do not have this dimord structure requirements. I'm puzzled why averaging over time and/or frequency could not be a valid option for actvsblT... could someone help me understand? Thanks! Akiko -- Akiko Ikkai, Ph.D. Postdoctoral Fellow Department of Psychological and Brain Sciences Johns Hopkins University Ames Hall, 3400 N. Charles St. Baltimore, MD 21218 -------------- next part -------------- An HTML attachment was scrubbed... URL: From aler1 at gmx.de Wed Aug 15 15:47:20 2012 From: aler1 at gmx.de (alla Brodski) Date: Wed, 15 Aug 2012 15:47:20 +0200 Subject: [FieldTrip] Source analysis power unit Message-ID: <20120815134720.43010@gmx.net> Dear fieldtrip users, I am using fieldtrip for beamforming source analysis. I try to interpret the units of the source power results, which are for my data in the range of 10^-07 to 10^-10. I am using the fieldtrip version of 01.05.2012; with an older version (e.g. of 2009) I get the same numbers but in a range of 10^27 to 10^30. I have already noticed that there was a change for the singleshell method, which I am using; a new scaling factor was introduced in the beginning of 2011: http://mailman.science.ru.nl/pipermail/fieldtrip/2011-January/003367.html However, I am still wondering whether I can interpret the results in physical units or not. Can they still be interpreted in tesla /fT (as if they had the old scale)? Or am I possibly doing something wrong as 10^-07 to 10^-10 does not seem a meaningful range. Thank you very much! Alla Brodski From aler1 at gmx.de Wed Aug 15 16:07:16 2012 From: aler1 at gmx.de (alla Brodski) Date: Wed, 15 Aug 2012 16:07:16 +0200 Subject: [FieldTrip] Source analysis power unit In-Reply-To: <20120815134720.43010@gmx.net> References: <20120815134720.43010@gmx.net> Message-ID: <20120815140716.43010@gmx.net> Sorry, of course it is 10^-27 to 10^-30 with an old version -------- Original-Nachricht -------- > Datum: Wed, 15 Aug 2012 15:47:20 +0200 > Von: "alla Brodski" > An: fieldtrip at science.ru.nl > Betreff: [FieldTrip] Source analysis power unit > Dear fieldtrip users, > > I am using fieldtrip for beamforming source analysis. > I try to interpret the units of the source power results, which are for my > data in the range of 10^-07 to 10^-10. I am using the fieldtrip version of > 01.05.2012; with an older version (e.g. of 2009) I get the same numbers > but in a range of 10^27 to 10^30. > I have already noticed that there was a change for the singleshell method, > which I am using; a new scaling factor was introduced in the beginning of > 2011: > http://mailman.science.ru.nl/pipermail/fieldtrip/2011-January/003367.html > > However, I am still wondering whether I can interpret the results in > physical units or not. Can they still be interpreted in tesla /fT (as if they > had the old scale)? Or am I possibly doing something wrong as 10^-07 to > 10^-10 does not seem a meaningful range. > > Thank you very much! > > Alla Brodski > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From Paul_C at gmx.de Fri Aug 17 01:12:04 2012 From: Paul_C at gmx.de (Paul Kertscher) Date: Fri, 17 Aug 2012 01:12:04 +0200 Subject: [FieldTrip] Question about ft_dipolesimulation Message-ID: <20120816231204.162220@gmx.net> Dear fieldtrippers, I recetly asked myself what units the signal in ft_dipolsimulation had. First I thought, that it had to be a current, thus Ampére. After rechecking it in the Kybic et al. paper it obviously is an current density. If the signal simulated has an amplitude of 1, does this mean, that the current density would be 1A/m², or are these units in the function rather of an arbitrary kind? Best regards, Paul Kertscher --- Paul Kertscher (Dipl.-Phys.) Björnsonstr. 25 12163 Berlin +49/(0)30/221609359 +49/(0)1578/7839933 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Andrew.Dimitrijevic at cchmc.org Fri Aug 17 04:41:40 2012 From: Andrew.Dimitrijevic at cchmc.org (Dimitrijevic, Andrew) Date: Fri, 17 Aug 2012 02:41:40 +0000 Subject: [FieldTrip] postdoc position available ... auditory ERPs in cochlear implant users Message-ID: <1787E0F7D5DAD54E803CBB4E877A33BE026B7C90@MCEXMB3.chmccorp.cchmc.org> Dear FieldTrippers ... postdoc position available, feel free to pass along (sorry for cross posting) Postdoctoral position in auditory EPs with cochlear implant users We are seeking a highly motivated postdoctoral fellow with expertise in EEGLAB/FieldTrip to the join the Hearing Lab at the Communication Sciences Research Center (CSRC) at Cincinnati Children’s Hospital Medical Center (CCHMC). This particular project will involve psychoacoustics and high-density EEG recordings in adults and children with cochlear implants. Much of the work will involve post processing in Matlab therefore solid programming ability is essential. The ideal candidate will have a PhD in cognitive science/neuroscience/computer science/psychology/engineering or related discipline, and an interest in cochlear implants and EEG. The CSRC has a strong interdisciplinary community that has both clinicians and basic scientists. Our facilities include an EEG lab, a Neuroimaging center with research dedicated 3T MRI scanner, 275-channel MEG lab. The position is for 2 years and second year is contingent on satisfactory progress. The anticipated start date is Jan 2012. Applicants should submit a cover letter and CV, including the names and contact information of three references. Please send application materials by e-mail to: Andrew Dimitrijevic PhD Assistant Professor Communication Sciences Research Center https://csrc.cchmc.org/ andrew.dimitrijevic at cchmc.org https://csrc.cchmc.org/andrew-dimitrijevic CCHMC has a longstanding commitment to achieving diversity among faculty, staff, and students. CCHMC is an EO/AA Employer. -------------- next part -------------- An HTML attachment was scrubbed... URL: From julian.keil at gmail.com Fri Aug 17 14:50:25 2012 From: julian.keil at gmail.com (Julian Keil) Date: Fri, 17 Aug 2012 14:50:25 +0200 Subject: [FieldTrip] FIeldTrip Tutorian Videos Message-ID: <82B32158-73E9-4E12-B91A-C398B996EB34@gmail.com> Dear FieldTrippers, a while ago I held a little workshop at the BRAMS Institute in Montreal covering the basic steps in FieldTrip. I recorded the whole session and finally got around to put the videos online. Please feel free to check it out at or point any new users to it: http://vimeo.com/user11934546/videos The accompanying Matlab script can be found here: http://pastebin.com/yskmVKAh Please also tell me if I made any errors or you feel that I have explained something incorrectly. Best, Julian ******************************************** Dr. Julian Keil International Laboratory for Brain Music and Sound Research (BRAMS) Pavillon 1430 Mont-Royal Université de Montréal Montréal, Québec Canada, H2V 4P3 julian.keil at gmail.com +1-514-343-6111-29653 www.brams.org -------------- next part -------------- An HTML attachment was scrubbed... URL: From pgoodin at swin.edu.au Sun Aug 19 13:17:16 2012 From: pgoodin at swin.edu.au (Peter Goodin) Date: Sun, 19 Aug 2012 11:17:16 +0000 Subject: [FieldTrip] Wavelet advice Message-ID: Hi Fieldtrip list, I'm attempting to use wavelets to analyse some data collected with a neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is 1.6 seconds) but am new to time frequency analysis. I'm interested in lower frequency bins (~4 to 20 Hz). The config settings I've been playing with are as follows: cfg.method = 'wavelet'; cfg.channel = 'MEG'; cfg.keeptrials = 'yes'; cfg.foi = [4:1:20] cfg.toi = [-.1:.01:1.5] cfg.width = x Using a default width of 7, I'm getting large boundary effects leading to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. I've played with the width and have found that a width of 1 gives only a small amount of boundary effect at the extreme edges and shows a large increase in the frequencies I'm interested in at a time point which corresponds quite nicely with the ERF data also analysed. Having read the Tallon-Baudry (1999) article, material from the tutorials and items from the mailing list, I understand that by using a wavelet width of 1 I've biased my results towards higher temporal vs. spectral resolution, but considering the low range of frequencies I'm interest in, is this a problem? Does using a low width with my cfg.foi set as it is just lead to a large amount of redundant data? Advice on this subject would be greatly appreciated. Peter. -------------- next part -------------- An HTML attachment was scrubbed... URL: From smoratti at psi.ucm.es Sun Aug 19 14:19:26 2012 From: smoratti at psi.ucm.es (smoratti at psi.ucm.es) Date: Sun, 19 Aug 2012 14:19:26 +0200 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: Message-ID: Dear Peter, I think you should use 5 cycles as minimum to get stable results. One cycle is not much to estimate the power of your time-freq of interest. Best, Stephan ________________________________________________________ Stephan Moratti, PhD see also: http://web.me.com/smoratti/ Universidad Complutense de Madrid Facultad de Psicología Departamento de Psicología Básica I Campus de Somosaguas 28223 Pozuelo de Alarcón (Madrid) Spain and Center for Biomedical Technology Laboratory for Cognitive and Computational Neuroscience Parque Científico y Tecnológico de la Universidad Politecnica de Madrid Campus Montegancedo 28223 Pozuelo de Alarcón (Madrid) Spain email: smoratti at psi.ucm.es Tel.: +34 679219982 El 19/08/2012, a las 13:17, Peter Goodin escribió: > Hi Fieldtrip list, > > I'm attempting to use wavelets to analyse some data collected with a neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is 1.6 seconds) but am new to time frequency analysis. I'm interested in lower frequency bins (~4 to 20 Hz). The config settings I've been playing with are as follows: > > cfg.method = 'wavelet'; > cfg.channel = 'MEG'; > cfg.keeptrials = 'yes'; > cfg.foi = [4:1:20] > cfg.toi = [-.1:.01:1.5] > cfg.width = x > > Using a default width of 7, I'm getting large boundary effects leading to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. I've played with the width and have found that a width of 1 gives only a small amount of boundary effect at the extreme edges and shows a large increase in the frequencies I'm interested in at a time point which corresponds quite nicely with the ERF data also analysed. > > Having read the Tallon-Baudry (1999) article, material from the tutorials and items from the mailing list, I understand that by using a wavelet width of 1 I've biased my results towards higher temporal vs. spectral resolution, but considering the low range of frequencies I'm interest in, is this a problem? Does using a low width with my cfg.foi set as it is just lead to a large amount of redundant data? > > Advice on this subject would be greatly appreciated. > > Peter. > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From sherrykhan78 at gmail.com Sun Aug 19 22:58:21 2012 From: sherrykhan78 at gmail.com (Sheraz Khan) Date: Sun, 19 Aug 2012 16:58:21 -0400 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: Message-ID: Hi, You can always try variables number of cycles may be start with 3 and then go to 5. Sheraz Martinos Center MGH/MIT/Harvard On Aug 19, 2012 7:17 AM, "Peter Goodin" wrote: > Hi Fieldtrip list, > > I'm attempting to use wavelets to analyse some data collected with a > neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is > 1.6 seconds) but am new to time frequency analysis. I'm interested in lower > frequency bins (~4 to 20 Hz). The config settings I've been playing with > are as follows: > > cfg.method = 'wavelet'; > cfg.channel = 'MEG'; > cfg.keeptrials = 'yes'; > cfg.foi = [4:1:20] > cfg.toi = [-.1:.01:1.5] > cfg.width = x > > Using a default width of 7, I'm getting large boundary effects leading > to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. > I've played with the width and have found that a width of 1 gives only a > small amount of boundary effect at the extreme edges and shows a large > increase in the frequencies I'm interested in at a time point which > corresponds quite nicely with the ERF data also analysed. > > Having read the Tallon-Baudry (1999) article, material from the > tutorials and items from the mailing list, I understand that by using a > wavelet width of 1 I've biased my results towards higher temporal vs. > spectral resolution, but considering the low range of frequencies I'm > interest in, is this a problem? Does using a low width with my cfg.foi set > as it is just lead to a large amount of redundant data? > > Advice on this subject would be greatly appreciated. > > Peter. > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pgoodin at swin.edu.au Mon Aug 20 01:20:49 2012 From: pgoodin at swin.edu.au (Peter Goodin) Date: Sun, 19 Aug 2012 23:20:49 +0000 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: , Message-ID: Thanks for all the replies so far. Re-reading my previous question I appear to have been a bit vague to the overall problem. The problem is I've used wavelets widths starting from 1 and increased to the default 7, but only a value of 1 gives me any usable data around 4Hz. What confuses me is that Kaplan et al, 2011 (http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001267) were able to get theta band activity using epochs of 1s (200ms pre 800ms post) without apparent boundary effects at a width of 5, but when I tried the same thing (including zero padding), the results were not usable. Again, any advice and explanations as to why 1. a value of 1 is not usable and 2. why I may be getting these boundary effects would be greatly appreciated. Peter ________________________________ From: fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl] on behalf of Sheraz Khan [sherrykhan78 at gmail.com] Sent: Monday, 20 August 2012 6:58 AM To: FieldTrip discussion list Subject: Re: [FieldTrip] Wavelet advice Hi, You can always try variables number of cycles may be start with 3 and then go to 5. Sheraz Martinos Center MGH/MIT/Harvard On Aug 19, 2012 7:17 AM, "Peter Goodin" > wrote: Hi Fieldtrip list, I'm attempting to use wavelets to analyse some data collected with a neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is 1.6 seconds) but am new to time frequency analysis. I'm interested in lower frequency bins (~4 to 20 Hz). The config settings I've been playing with are as follows: cfg.method = 'wavelet'; cfg.channel = 'MEG'; cfg.keeptrials = 'yes'; cfg.foi = [4:1:20] cfg.toi = [-.1:.01:1.5] cfg.width = x Using a default width of 7, I'm getting large boundary effects leading to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. I've played with the width and have found that a width of 1 gives only a small amount of boundary effect at the extreme edges and shows a large increase in the frequencies I'm interested in at a time point which corresponds quite nicely with the ERF data also analysed. Having read the Tallon-Baudry (1999) article, material from the tutorials and items from the mailing list, I understand that by using a wavelet width of 1 I've biased my results towards higher temporal vs. spectral resolution, but considering the low range of frequencies I'm interest in, is this a problem? Does using a low width with my cfg.foi set as it is just lead to a large amount of redundant data? Advice on this subject would be greatly appreciated. Peter. _______________________________________________ fieldtrip mailing list fieldtrip at donders.ru.nl http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From sherrykhan78 at gmail.com Mon Aug 20 03:29:42 2012 From: sherrykhan78 at gmail.com (Sheraz Khan) Date: Sun, 19 Aug 2012 21:29:42 -0400 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: Message-ID: You can not avoid the border effects they comes from the convolution. Only possible solution which I sometimes use when the epochs are shorter due to the design, is to do the wavelet decomposition on the continuous data, and then epoch it, this will avoid the borders from the epoch. Sheraz Martinos Center MGH/MIT/Harvard On Sun, Aug 19, 2012 at 7:20 PM, Peter Goodin wrote: > Thanks for all the replies so far. > > Re-reading my previous question I appear to have been a bit vague to the > overall problem. > > The problem is I've used wavelets widths starting from 1 and increased > to the default 7, but only a value of 1 gives me any usable data around > 4Hz. > > What confuses me is that Kaplan et al, 2011 ( > http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001267) > were able to get theta band activity using epochs of 1s (200ms pre 800ms > post) without apparent boundary effects at a width of 5, but when I tried > the same thing (including zero padding), the results were not usable. > > Again, any advice and explanations as to why 1. a value of 1 is not > usable and 2. why I may be getting these boundary effects would be greatly > appreciated. > > Peter > ------------------------------ > *From:* fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl] > on behalf of Sheraz Khan [sherrykhan78 at gmail.com] > *Sent:* Monday, 20 August 2012 6:58 AM > *To:* FieldTrip discussion list > *Subject:* Re: [FieldTrip] Wavelet advice > > Hi, > You can always try variables number of cycles may be start with 3 and then > go to 5. > Sheraz > Martinos Center > MGH/MIT/Harvard > On Aug 19, 2012 7:17 AM, "Peter Goodin" wrote: > >> Hi Fieldtrip list, >> >> I'm attempting to use wavelets to analyse some data collected with a >> neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is >> 1.6 seconds) but am new to time frequency analysis. I'm interested in lower >> frequency bins (~4 to 20 Hz). The config settings I've been playing with >> are as follows: >> >> cfg.method = 'wavelet'; >> cfg.channel = 'MEG'; >> cfg.keeptrials = 'yes'; >> cfg.foi = [4:1:20] >> cfg.toi = [-.1:.01:1.5] >> cfg.width = x >> >> Using a default width of 7, I'm getting large boundary effects leading >> to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. >> I've played with the width and have found that a width of 1 gives only a >> small amount of boundary effect at the extreme edges and shows a large >> increase in the frequencies I'm interested in at a time point which >> corresponds quite nicely with the ERF data also analysed. >> >> Having read the Tallon-Baudry (1999) article, material from the >> tutorials and items from the mailing list, I understand that by using a >> wavelet width of 1 I've biased my results towards higher temporal vs. >> spectral resolution, but considering the low range of frequencies I'm >> interest in, is this a problem? Does using a low width with my cfg.foi set >> as it is just lead to a large amount of redundant data? >> >> Advice on this subject would be greatly appreciated. >> >> Peter. >> >> >> >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip >> > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pgoodin at swin.edu.au Mon Aug 20 05:18:09 2012 From: pgoodin at swin.edu.au (Peter Goodin) Date: Mon, 20 Aug 2012 03:18:09 +0000 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: , Message-ID: Thanks for that! Much appreciated. ________________________________ From: fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl] on behalf of Sheraz Khan [sherrykhan78 at gmail.com] Sent: Monday, 20 August 2012 11:29 AM To: FieldTrip discussion list Subject: Re: [FieldTrip] Wavelet advice You can not avoid the border effects they comes from the convolution. Only possible solution which I sometimes use when the epochs are shorter due to the design, is to do the wavelet decomposition on the continuous data, and then epoch it, this will avoid the borders from the epoch. Sheraz Martinos Center MGH/MIT/Harvard On Sun, Aug 19, 2012 at 7:20 PM, Peter Goodin > wrote: Thanks for all the replies so far. Re-reading my previous question I appear to have been a bit vague to the overall problem. The problem is I've used wavelets widths starting from 1 and increased to the default 7, but only a value of 1 gives me any usable data around 4Hz. What confuses me is that Kaplan et al, 2011 (http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001267) were able to get theta band activity using epochs of 1s (200ms pre 800ms post) without apparent boundary effects at a width of 5, but when I tried the same thing (including zero padding), the results were not usable. Again, any advice and explanations as to why 1. a value of 1 is not usable and 2. why I may be getting these boundary effects would be greatly appreciated. Peter ________________________________ From: fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl] on behalf of Sheraz Khan [sherrykhan78 at gmail.com] Sent: Monday, 20 August 2012 6:58 AM To: FieldTrip discussion list Subject: Re: [FieldTrip] Wavelet advice Hi, You can always try variables number of cycles may be start with 3 and then go to 5. Sheraz Martinos Center MGH/MIT/Harvard On Aug 19, 2012 7:17 AM, "Peter Goodin" > wrote: Hi Fieldtrip list, I'm attempting to use wavelets to analyse some data collected with a neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is 1.6 seconds) but am new to time frequency analysis. I'm interested in lower frequency bins (~4 to 20 Hz). The config settings I've been playing with are as follows: cfg.method = 'wavelet'; cfg.channel = 'MEG'; cfg.keeptrials = 'yes'; cfg.foi = [4:1:20] cfg.toi = [-.1:.01:1.5] cfg.width = x Using a default width of 7, I'm getting large boundary effects leading to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. I've played with the width and have found that a width of 1 gives only a small amount of boundary effect at the extreme edges and shows a large increase in the frequencies I'm interested in at a time point which corresponds quite nicely with the ERF data also analysed. Having read the Tallon-Baudry (1999) article, material from the tutorials and items from the mailing list, I understand that by using a wavelet width of 1 I've biased my results towards higher temporal vs. spectral resolution, but considering the low range of frequencies I'm interest in, is this a problem? Does using a low width with my cfg.foi set as it is just lead to a large amount of redundant data? Advice on this subject would be greatly appreciated. Peter. _______________________________________________ fieldtrip mailing list fieldtrip at donders.ru.nl http://mailman.science.ru.nl/mailman/listinfo/fieldtrip _______________________________________________ fieldtrip mailing list fieldtrip at donders.ru.nl http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From nuria.donamayor at neuro.uni-luebeck.de Tue Aug 21 12:50:56 2012 From: nuria.donamayor at neuro.uni-luebeck.de (=?iso-8859-1?Q?Nuria_Do=F1amayor_Alonso?=) Date: Tue, 21 Aug 2012 12:50:56 +0200 Subject: [FieldTrip] source reconstruction on emfs Message-ID: <810A8E06C75EB447A8CEB73DBFD7BB0E4C03CD65AD@solaris.neuro.uni-luebeck.de> Dear fieldtrippers, I am currently analyzing a study, measured with a Neuromag 306 system, in which we stimulated the median nerve. I had never done source reconstruction with fieldtrip, so I followed the tutorial using MNE for EMF source analysis (which worked fine), only to find out that I could neither average individual sources nor perform stats, which is exactly what I need to do... So the question is: What would you guys recommend? Should I just try using the LCMV beamfomer? However, that might not be the best solution according to this thread http://mailman.science.ru.nl/pipermail/fieldtrip/2011-September/004321.html... Is there any way I could efficiently export my preprocessed/averaged data to some other software (MNE/SPM/Brainstorm...) and compute my source reconstruction and stats there? What do you guys usually do? Thanks a lot for your help, Nuria From jm.horschig at donders.ru.nl Tue Aug 21 13:24:21 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Tue, 21 Aug 2012 13:24:21 +0200 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: , Message-ID: <50336FE5.30602@donders.ru.nl> Hey Peter, when doing what Sheraz suggested you should, however, keep in mind that the edges are computed using data outside your epoch, i.e. possible muscle artifacts and other stuff (e.g. stimulation) will corrupt the frequency estimation if not properly accounted for. Make sure to note that when interpreting your data. Best, Jörn On 8/20/2012 5:18 AM, Peter Goodin wrote: > Thanks for that! Much appreciated. > ------------------------------------------------------------------------ > *From:* fieldtrip-bounces at science.ru.nl > [fieldtrip-bounces at science.ru.nl] on behalf of Sheraz Khan > [sherrykhan78 at gmail.com] > *Sent:* Monday, 20 August 2012 11:29 AM > *To:* FieldTrip discussion list > *Subject:* Re: [FieldTrip] Wavelet advice > > You can not avoid the border effects they comes from the convolution. > > Only possible solution which I sometimes use when the epochs are > shorter due to the design, is to do the wavelet decomposition on > the continuous data, and then epoch it, this will avoid the borders > from the epoch. > > Sheraz > Martinos Center > MGH/MIT/Harvard > > On Sun, Aug 19, 2012 at 7:20 PM, Peter Goodin > wrote: > > Thanks for all the replies so far. > > Re-reading my previous question I appear to have been a bit vague > to the overall problem. > > The problem is I've used wavelets widths starting from 1 and > increased to the default 7, but only a value of 1 gives me any > usable data around 4Hz. > > What confuses me is that Kaplan et al, 2011 > (http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001267) > were able to get theta band activity using epochs of 1s (200ms pre > 800ms post) without apparent boundary effects at a width of 5, but > when I tried the same thing (including zero padding), the results > were not usable. > > Again, any advice and explanations as to why 1. a value of 1 is > not usable and 2. why I may be getting these boundary effects > would be greatly appreciated. > > Peter > ------------------------------------------------------------------------ > *From:* fieldtrip-bounces at science.ru.nl > > [fieldtrip-bounces at science.ru.nl > ] on behalf of Sheraz Khan > [sherrykhan78 at gmail.com ] > *Sent:* Monday, 20 August 2012 6:58 AM > *To:* FieldTrip discussion list > *Subject:* Re: [FieldTrip] Wavelet advice > > Hi, > You can always try variables number of cycles may be start with 3 > and then go to 5. > Sheraz > Martinos Center > MGH/MIT/Harvard > > On Aug 19, 2012 7:17 AM, "Peter Goodin" > wrote: > > Hi Fieldtrip list, > > I'm attempting to use wavelets to analyse some data collected > with a neuromag system (sample rate of 500Hz, pre-trigger > period is 200ms, post is 1.6 seconds) but am new to time > frequency analysis. I'm interested in lower frequency bins (~4 > to 20 Hz). The config settings I've been playing with are as > follows: > > cfg.method = 'wavelet'; > cfg.channel = 'MEG'; > cfg.keeptrials = 'yes'; > cfg.foi = [4:1:20] > cfg.toi = [-.1:.01:1.5] > cfg.width = x > > Using a default width of 7, I'm getting large boundary effects > leading to a period of ~300ms (400 - 700ms post trigger) of > calculated data at 4Hz. I've played with the width and have > found that a width of 1 gives only a small amount of boundary > effect at the extreme edges and shows a large increase in the > frequencies I'm interested in at a time point which > corresponds quite nicely with the ERF data also analysed. > > Having read the Tallon-Baudry (1999) article, material from > the tutorials and items from the mailing list, I understand > that by using a wavelet width of 1 I've biased my results > towards higher temporal vs. spectral resolution, but > considering the low range of frequencies I'm interest in, is > this a problem? Does using a low width with my cfg.foi set as > it is just lead to a large amount of redundant data? > > Advice on this subject would be greatly appreciated. > > Peter. > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group FieldTrip Development Team P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From r.oostenveld at donders.ru.nl Tue Aug 21 21:56:20 2012 From: r.oostenveld at donders.ru.nl (Robert Oostenveld) Date: Tue, 21 Aug 2012 21:56:20 +0200 Subject: [FieldTrip] source reconstruction on emfs In-Reply-To: <810A8E06C75EB447A8CEB73DBFD7BB0E4C03CD65AD@solaris.neuro.uni-luebeck.de> References: <810A8E06C75EB447A8CEB73DBFD7BB0E4C03CD65AD@solaris.neuro.uni-luebeck.de> Message-ID: Dear Nuria For the earliest components from S1, and just to get started and get a feel for your data, a traditional dipole fit using FT_DIPOLEFITTING should work fine. Following the initial localization of S1, you indeed could continue with an LCMV beamformer. There is an example at http://fieldtrip.fcdonders.nl/example/fit_a_dipole_to_the_tactile_erf_after_mechanical_stimulation The example is on CTF data with Braille-cell stimulation (i.e. mechanical instead of electrical),. Since the example data is available from the ftp you can try the example out before translating it to your own data. Furthermore I should note that the example is rather old and probably has not been run for quite some time, so there might be some pieces of the example script on the wiki that need updating to reflect the current state of the fieldtrip code. I suggest you also have a look here http://fieldtrip.fcdonders.nl/tutorial/headmodel_meg, which is a new tutorial that Lilla Magyari (CC) made available last week. best regards Robert On 21 Aug 2012, at 12:50, Nuria Doñamayor Alonso wrote: > Dear fieldtrippers, > I am currently analyzing a study, measured with a Neuromag 306 system, in which we stimulated the median nerve. I had never done source reconstruction with fieldtrip, so I followed the tutorial using MNE for EMF source analysis (which worked fine), only to find out that I could neither average individual sources nor perform stats, which is exactly what I need to do... > So the question is: What would you guys recommend? Should I just try using the LCMV beamfomer? However, that might not be the best solution according to this thread http://mailman.science.ru.nl/pipermail/fieldtrip/2011-September/004321.html... Is there any way I could efficiently export my preprocessed/averaged data to some other software (MNE/SPM/Brainstorm...) and compute my source reconstruction and stats there? What do you guys usually do? > Thanks a lot for your help, > Nuria > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From Hisako.Fujiwara at cchmc.org Tue Aug 21 22:13:04 2012 From: Hisako.Fujiwara at cchmc.org (Fujiwara, Hisako) Date: Tue, 21 Aug 2012 20:13:04 +0000 Subject: [FieldTrip] beamformer application to spontaneous signal source construction Message-ID: Dear all, I have a spontaneous recording of a patient who has epilepsy. I want to use beamformer to reconstruct and display an increase power at certain frequency and location in the brain. Your example only describe how to use beamfomer for ERP. Could you kindly advise how to apply your fieldtrip beamformer method for my needed application. I would like to thank your all kind help and advice in advance. Sincerely, Hisako -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Wed Aug 22 19:59:51 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Wed, 22 Aug 2012 19:59:51 +0200 Subject: [FieldTrip] FT_REJECTARTIFACT Message-ID: Dear all, I have one problem concerning ft_rejectartifact function. I did everything as you explained in automatic artifact rejection tutorial. I mark segments first and that is fine, but the ft_rejectartifact doesn't remove marked bad segments form my data. Please help! Here is my code: % automatic clear all s=[87 209 195 225 60]; for g=1:length(s) input_no=s(g) clear sub sub=['Subject' num2str(input_no)]; eval(sub) %calling subject's script cfg=[]; cfg.trialdef.eventtype = 'UPPT001'; cfg.trialdef.eventvalue = 1; cfg.trialdef.prestim = 0.1; cfg.trialdef.poststim = 0.5; cfg.trialfun = 'trialfun_general'; cfg.channel = {'MEG'}; cfg.continuous = 'yes'; cfg.dataset = [subjectdata.wr11.datadir]; [data1]=ft_definetrial(cfg) trial=data1.trl; % muscle cfg=[]; cfg.trl =trial; cfg.datafile = [subjectdata.wr11.datafile]; cfg.headerfile =[subjectdata.wr11.headerfile]; cfg.continuous = 'yes'; % channel selection, cutoff and padding cfg.artfctdef.zvalue.channel = 'MRT*'; cfg.artfctdef.zvalue.cutoff = 15; cfg.artfctdef.zvalue.trlpadding = 0.5; cfg.artfctdef.zvalue.fltpadding = 0.5; cfg.artfctdef.zvalue.artpadding = 0.1; % algorithmic parameters cfg.artfctdef.zvalue.bpfilter = 'yes'; cfg.artfctdef.zvalue.bpfreq = [110 140]; cfg.artfctdef.zvalue.bpfiltord = 9; cfg.artfctdef.zvalue.bpfilttype = 'but'; cfg.artfctdef.zvalue.hilbert = 'yes'; cfg.artfctdef.zvalue.boxcar = 0.2; %make the process interactive cfg.artfctdef.zvalue.interactive = 'yes'; [cfg, artifact_muscle] = ft_artifact_zvalue(cfg) % jump cfg=[]; cfg.trl = trial; cfg.datafile = [subjectdata.wr11.datafile]; cfg.headerfile =[subjectdata.wr11.headerfile]; cfg.continuous = 'yes'; % channel selection, cutoff and padding cfg.artfctdef.zvalue.channel = 'MEG'; cfg.artfctdef.zvalue.cutoff = 20; cfg.artfctdef.zvalue.trlpadding = 0; cfg.artfctdef.zvalue.artpadding = 0; cfg.artfctdef.zvalue.fltpadding = 0; % algorithmic parameters cfg.artfctdef.zvalue.cumulative = 'yes'; cfg.artfctdef.zvalue.medianfilter = 'yes'; cfg.artfctdef.zvalue.medianfiltord = 9; cfg.artfctdef.zvalue.absdiff = 'yes'; % make the process interactive cfg.artfctdef.zvalue.interactive = 'yes'; [cfg, artifact_jump] = ft_artifact_zvalue(cfg) cfg.artfctdef.reject = 'complete'; cfg.artfctdef.jump.artifact = artifact_jump; cfg.artfctdef.muscle.artifact = artifact_muscle; data_no_artifacts = ft_rejectartifact(data1) end Thank you in advance! NP -------------- next part -------------- An HTML attachment was scrubbed... URL: From eelke.spaak at donders.ru.nl Wed Aug 22 20:17:29 2012 From: eelke.spaak at donders.ru.nl (Eelke Spaak) Date: Wed, 22 Aug 2012 20:17:29 +0200 Subject: [FieldTrip] FT_REJECTARTIFACT In-Reply-To: References: Message-ID: Dear Nenad, My guess is you are missing the first argument in your call to ft_rejectartifact; i.e. instead of data_no_artifacts = ft_rejectartifact(data1) you should use data_no_artifacts = ft_rejectartifact(cfg, data1) Does that work? Best, Eelke On 22 August 2012 19:59, Nenad Polomac wrote: > Dear all, > > I have one problem concerning ft_rejectartifact function. > I did everything as you explained in automatic artifact rejection tutorial. > I mark segments first and that is fine, but the ft_rejectartifact doesn't > remove marked bad segments form my data. > Please help! > > Here is my code: > > % automatic > clear all > > s=[87 209 195 225 60]; > > > for g=1:length(s) > input_no=s(g) > > clear sub > > sub=['Subject' num2str(input_no)]; > eval(sub) %calling subject's script > > > cfg=[]; > cfg.trialdef.eventtype = 'UPPT001'; > cfg.trialdef.eventvalue = 1; > cfg.trialdef.prestim = 0.1; > cfg.trialdef.poststim = 0.5; > cfg.trialfun = 'trialfun_general'; > cfg.channel = {'MEG'}; > cfg.continuous = 'yes'; > cfg.dataset = [subjectdata.wr11.datadir]; > > [data1]=ft_definetrial(cfg) > trial=data1.trl; > > > % muscle > cfg=[]; > cfg.trl =trial; > cfg.datafile = [subjectdata.wr11.datafile]; > cfg.headerfile =[subjectdata.wr11.headerfile]; > cfg.continuous = 'yes'; > > % channel selection, cutoff and padding > cfg.artfctdef.zvalue.channel = 'MRT*'; > cfg.artfctdef.zvalue.cutoff = 15; > cfg.artfctdef.zvalue.trlpadding = 0.5; > cfg.artfctdef.zvalue.fltpadding = 0.5; > cfg.artfctdef.zvalue.artpadding = 0.1; > > % algorithmic parameters > cfg.artfctdef.zvalue.bpfilter = 'yes'; > cfg.artfctdef.zvalue.bpfreq = [110 140]; > cfg.artfctdef.zvalue.bpfiltord = 9; > cfg.artfctdef.zvalue.bpfilttype = 'but'; > cfg.artfctdef.zvalue.hilbert = 'yes'; > cfg.artfctdef.zvalue.boxcar = 0.2; > > %make the process interactive > cfg.artfctdef.zvalue.interactive = 'yes'; > > [cfg, artifact_muscle] = ft_artifact_zvalue(cfg) > > > > % jump > cfg=[]; > cfg.trl = trial; > cfg.datafile = [subjectdata.wr11.datafile]; > cfg.headerfile =[subjectdata.wr11.headerfile]; > cfg.continuous = 'yes'; > > % channel selection, cutoff and padding > cfg.artfctdef.zvalue.channel = 'MEG'; > cfg.artfctdef.zvalue.cutoff = 20; > cfg.artfctdef.zvalue.trlpadding = 0; > cfg.artfctdef.zvalue.artpadding = 0; > cfg.artfctdef.zvalue.fltpadding = 0; > > % algorithmic parameters > cfg.artfctdef.zvalue.cumulative = 'yes'; > cfg.artfctdef.zvalue.medianfilter = 'yes'; > cfg.artfctdef.zvalue.medianfiltord = 9; > cfg.artfctdef.zvalue.absdiff = 'yes'; > > % make the process interactive > cfg.artfctdef.zvalue.interactive = 'yes'; > > [cfg, artifact_jump] = ft_artifact_zvalue(cfg) > > > > > cfg.artfctdef.reject = 'complete'; > cfg.artfctdef.jump.artifact = artifact_jump; > cfg.artfctdef.muscle.artifact = artifact_muscle; > data_no_artifacts = ft_rejectartifact(data1) > end > > Thank you in advance! > NP > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From polomacnenad at gmail.com Thu Aug 23 10:50:20 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Thu, 23 Aug 2012 10:50:20 +0200 Subject: [FieldTrip] FT_REJECTARTIFACT Message-ID: Dear Eelke, Thank you for your suggestion. But unfortunately this doesn't work. If I do your way I get: Error using ft_rejectartifact (line 233) no trials were selected, cannot perform artifact detection/rejection. So I am still not sure what might be the problem... Regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From stan.vanpelt at fcdonders.ru.nl Thu Aug 23 11:25:21 2012 From: stan.vanpelt at fcdonders.ru.nl (Stan van Pelt) Date: Thu, 23 Aug 2012 11:25:21 +0200 (CEST) Subject: [FieldTrip] FT_REJECTARTIFACT In-Reply-To: Message-ID: <1052918099.53046.1345713921693.JavaMail.root@sculptor.zimbra.ru.nl> Dear Nenad, It seems that no trials are defined when you run ft_definetrial. Can you check if there are actually is trial data present in your data1 structure? If there is no data, you should check if your configuration settings are correct in that processing step (e.g., trigger channel, event values). Best, Stan ----- Oorspronkelijk bericht ----- > Van: "Nenad Polomac" > Aan: fieldtrip at science.ru.nl > Verzonden: Donderdag 23 augustus 2012 10:50:20 > Onderwerp: Re: [FieldTrip] FT_REJECTARTIFACT > Dear Eelke, > Thank you for your suggestion. But unfortunately this doesn't work. If > I do your way I get: Error using ft_rejectartifact (line 233) > no trials were selected, cannot perform artifact detection/rejection. > So I am still not sure what might be the problem... > Regards! > Nenad > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Stan van Pelt Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen Kapittelweg 29, 6525 EN Nijmegen, Netherlands E-mail: stan.vanpelt at donders.ru.nl Website: www.ru.nl/donders/ Phone: (+31) (0)24 36 10981 Fax: (+31) (0)24 36 10989 -------------- next part -------------- An HTML attachment was scrubbed... URL: From poortjasper at gmail.com Thu Aug 23 12:34:48 2012 From: poortjasper at gmail.com (Jasper Poort) Date: Thu, 23 Aug 2012 11:34:48 +0100 Subject: [FieldTrip] Fwd: ft_freqdescriptives and cfg.trial In-Reply-To: References: Message-ID: Dear Fieldtrip users, I have a problem with using a trial selection to do freqdescriptives on a subset of trials: I first do ft_freqanalysis with these setttings, which results in freq: cfg.output = 'fourier'; cfg.method = 'mtmconvol'; cfg.keeptrials = 'yes'; freq = label: {1x64 cell} dimord: 'rpttap_chan_freq_time' freq: [30 35 40 45 50 55 60 65 70 75 80 85 90 95 100] time: [1x36 double] fourierspctrm: [4-D double] cumtapcnt: [857x15 double] cfg: [1x1 struct] size(freq.fourierspctrm) ans = 2571 64 15 36 then i try running ft_freqdescriptives with these settings tmpcfg = []; tmpcfg.jackknife = 'no'; tmpcfg.keeptrials = 'no'; tmpcfg.channel = freq.label([chnix1,chnix2]) tmpcfg.trials = [1:10:100]; tmp = ft_freqdescriptives(tmpcfg,freq); Error using + Array dimensions must match for binary array op. Error in ft_checkdata>fixcsd (line 740) powspctrm = powspctrm + abs(data.fourierspctrm(p:ntap:end,:,:,:,:)).^2; Error in ft_checkdata (line 646) data = fixcsd(data, cmbrepresentation, channelcmb); Error in ft_freqdescriptives (line 131) freq = ft_checkdata(freq, 'cmbrepresentation', 'sparsewithpow', 'channelcmb', {}); the error seems to happen because the cumtapcnt seems to be updated after calling ft_selectdata in ft_freqdescriptives such that it no longer contains a row for every trial which causes nrpt = size(data.cumtapcnt,1); in ft_checkdata to return 1 instead of the number of trials. label: {2x1 cell} dimord: 'rpttap_chan_freq_time' freq: [30 35 40 45 50 55 60 65 70 75 80 85 90 95 100] time: [1x36 double] fourierspctrm: [4-D double] cumtapcnt: [3 3 3 3 3 3 3 3 3 3] cfg: [1x1 struct] I was wondering if this is a bug or am I somehow specifying the input incorrectly? Thanks! best, Jasper -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 23 12:55:02 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 23 Aug 2012 12:55:02 +0200 Subject: [FieldTrip] Fwd: ft_freqdescriptives and cfg.trial In-Reply-To: References: Message-ID: Hi Jasper, I think this is indeed a bug in fieldtrip. The reason being that the cumtapcnt field for timefrequency data is not well-defined/well-designed/well-handled (the latter by ft_selectdata, which is doing the trial selection). Probably you can work around this by updating the cumtapcnt field prior to calling ft_freqdescriptives. I suspect that it will work if you define it to be a vector (nx1) where n is the number of trials, and each entry has the value of the number of tapers applied (per trial). Would it be possible for you to create a bug on our bugzilla site (bugzilla.fcdonders.nl) with a description of the problem, a little test script and some (small) data for reproduction? The team will then take it from there. Best, Jan-Mathijs On Aug 23, 2012, at 12:34 PM, Jasper Poort wrote: > Dear Fieldtrip users, > > > > I have a problem with using a trial selection to do freqdescriptives on a subset of trials: > > > > I first do ft_freqanalysis with these setttings, which results in freq: > > > > cfg.output = 'fourier'; > > cfg.method = 'mtmconvol'; > > cfg.keeptrials = 'yes'; > > > freq = > > > label: {1x64 cell} > > dimord: 'rpttap_chan_freq_time' > > freq: [30 35 40 45 50 55 60 65 70 75 80 85 90 95 100] > > time: [1x36 double] > > fourierspctrm: [4-D double] > > cumtapcnt: [857x15 double] > > cfg: [1x1 struct] > > > size(freq.fourierspctrm) > > ans = > > 2571 64 15 36 > > > then i try running ft_freqdescriptives with these settings > > tmpcfg = []; > > tmpcfg.jackknife = 'no'; > > tmpcfg.keeptrials = 'no'; > > tmpcfg.channel = freq.label([chnix1,chnix2]) > > tmpcfg.trials = [1:10:100]; > > tmp = ft_freqdescriptives(tmpcfg,freq); > > > Error using + > > Array dimensions must match for binary array op. > > > Error in ft_checkdata>fixcsd (line 740) > > powspctrm = powspctrm + abs(data.fourierspctrm(p:ntap:end,:,:,:,:)).^2; > > > Error in ft_checkdata (line 646) > > data = fixcsd(data, cmbrepresentation, channelcmb); > > > Error in ft_freqdescriptives (line 131) > > freq = ft_checkdata(freq, 'cmbrepresentation', 'sparsewithpow', 'channelcmb', {}); > > > > the error seems to happen because the cumtapcnt seems to be updated after calling ft_selectdata in ft_freqdescriptives such that it no longer contains a row for every trial which causes nrpt = size(data.cumtapcnt,1); in ft_checkdata to return 1 instead of the number of trials. > > > label: {2x1 cell} > dimord: 'rpttap_chan_freq_time' > freq: [30 35 40 45 50 55 60 65 70 75 80 85 90 95 100] > time: [1x36 double] > fourierspctrm: [4-D double] > cumtapcnt: [3 3 3 3 3 3 3 3 3 3] > cfg: [1x1 struct] > > > > I was wondering if this is a bug or am I somehow specifying the input incorrectly? > > > > Thanks! best, Jasper > > > > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jm.horschig at donders.ru.nl Thu Aug 23 13:41:00 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Thu, 23 Aug 2012 13:41:00 +0200 Subject: [FieldTrip] FT_REJECTARTIFACT In-Reply-To: References: Message-ID: <503616CC.7010303@donders.ru.nl> Dear Nenad, it is a bit confusing that you call the variable returned from ft_definetrial 'data1', because actually it is only a cfg-structure without data. No matter how you call that variable though, it should work if you use this: / data1.artfctdef.reject = 'complete'; / / data1.artfctdef.jump.artifact = artifact_jump;/ / data1.artfctdef.muscle.artifact = artifact_muscle;/ / data_no_artifacts = ft_rejectartifact(data1)/ Note that you need to call ft_preprocessing in order to obtain data, in your case: /data = ft_preprocessing(data_no_artifacts);/ or /data = ft_preprocessing(data1);/ ft_definetrial is a function which just returns at what samples your trials are supposed to start. ft_preprocessing is then using this information to extract the trials out of your datafile. You can artifact rejection either before you segmented the data into trials, as done here, or afterwards, i.e. after having called ft_preprocessing. Best, Jörn On 8/23/2012 10:50 AM, Nenad Polomac wrote: > Dear Eelke, > > Thank you for your suggestion. But unfortunately this doesn't work. If > I do your way I get: Error using ft_rejectartifact (line 233) > no trials were selected, cannot perform artifact detection/rejection. > > So I am still not sure what might be the problem... > > Regards! > > Nenad > > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group FieldTrip Development Team P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Thu Aug 23 14:46:37 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Thu, 23 Aug 2012 14:46:37 +0200 Subject: [FieldTrip] FT_REJECTARTIFACT In-Reply-To: References: Message-ID: Dear Jörn, Thank you for your answer! That was the problem. I understand all now. Regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From s.rombetto at cib.na.cnr.it Thu Aug 23 14:57:51 2012 From: s.rombetto at cib.na.cnr.it (s.rombetto at cib.na.cnr.it) Date: Thu, 23 Aug 2012 14:57:51 +0200 Subject: [FieldTrip] EDF data Message-ID: <20120823145751.avcc2gkcw0ws0wow@arco.cib.na.cnr.it> Dear fieldtrip users, I'm trying to import some EEG data in EDF format. I am able to read the data structure using hdr= ft_read_header(rawdataname); and also to read the data by data = ft_read_data(rawdataname);, but I receive an error message when I try to perform the ft_preprocessing command (Error in ==> ft_preprocessing at 511). Do you have any hint or sample code? Thank you in advance! Sara ------------------------- Dott.ssa Sara Rombetto Istituto di Cibernetica "E. Caianiello" Via Campi Flegrei, 34 80078 Pozzuoli (NA) Italy tel +390818675361 fax +390818675128 -------------------------- "I disapprove of what you say, but I will defend to the death your right to say it." [Evelyn Beatrice Hall, The Friends Of Voltaire] ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From nuria.donamayor at neuro.uni-luebeck.de Thu Aug 23 16:03:04 2012 From: nuria.donamayor at neuro.uni-luebeck.de (=?iso-8859-1?Q?Nuria_Do=F1amayor_Alonso?=) Date: Thu, 23 Aug 2012 16:03:04 +0200 Subject: [FieldTrip] source reconstruction on emfs In-Reply-To: References: <810A8E06C75EB447A8CEB73DBFD7BB0E4C03CD65AD@solaris.neuro.uni-luebeck.de>, Message-ID: <810A8E06C75EB447A8CEB73DBFD7BB0E4C03CD65C1@solaris.neuro.uni-luebeck.de> Dear Robert, I did what you suggested and everything worked perfectly, thanks. Now I have a slight problem with the following. I want to delimit a ROI to do stats on my LCMV results and, as far as I understand, I should use ft_sourceinterpolate, ft_volumenormalise and then ft_sourcegrandaverage. The problem is, when I try to check out the result with ft_sourceplot, it just plots the first slice 20 times. This does not happen with the non-normalised grandaverages (and the individual averages in both cases). Any suggestions? Here's the code I'm using: # on the individual data cfg = []; cdf.downsample = 2; cfg.parameter = 'avg.nai'; sourceint = ft_sourceinterpolate(cfg, source, mri); cfg = []; cfg.coordsys = 'neuromag'; cfg.nonlinear = 'no'; sourcenorm = ft_volumenormalise(cfg, sourceint); # grandaverage cfg = []; cfg.keepindividual = 'yes'; cfg.parameter = 'nai'; gsource = (cfg, sourceint1, sourceint2...); # plot gsource cfg = []; cfg.funparameter = 'avg.nai'; cfg.maskparameter = cfg.funparameter; ft_sourceplot(cfg, gsource); Thanks again! Nuria ________________________________________ Von: fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl] im Auftrag von Robert Oostenveld [r.oostenveld at donders.ru.nl] Gesendet: Dienstag, 21. August 2012 21:56 An: FieldTrip discussion list Betreff: Re: [FieldTrip] source reconstruction on emfs Dear Nuria For the earliest components from S1, and just to get started and get a feel for your data, a traditional dipole fit using FT_DIPOLEFITTING should work fine. Following the initial localization of S1, you indeed could continue with an LCMV beamformer. There is an example at http://fieldtrip.fcdonders.nl/example/fit_a_dipole_to_the_tactile_erf_after_mechanical_stimulation The example is on CTF data with Braille-cell stimulation (i.e. mechanical instead of electrical),. Since the example data is available from the ftp you can try the example out before translating it to your own data. Furthermore I should note that the example is rather old and probably has not been run for quite some time, so there might be some pieces of the example script on the wiki that need updating to reflect the current state of the fieldtrip code. I suggest you also have a look here http://fieldtrip.fcdonders.nl/tutorial/headmodel_meg, which is a new tutorial that Lilla Magyari (CC) made available last week. best regards Robert On 21 Aug 2012, at 12:50, Nuria Doñamayor Alonso wrote: > Dear fieldtrippers, > I am currently analyzing a study, measured with a Neuromag 306 system, in which we stimulated the median nerve. I had never done source reconstruction with fieldtrip, so I followed the tutorial using MNE for EMF source analysis (which worked fine), only to find out that I could neither average individual sources nor perform stats, which is exactly what I need to do... > So the question is: What would you guys recommend? Should I just try using the LCMV beamfomer? However, that might not be the best solution according to this thread http://mailman.science.ru.nl/pipermail/fieldtrip/2011-September/004321.html... Is there any way I could efficiently export my preprocessed/averaged data to some other software (MNE/SPM/Brainstorm...) and compute my source reconstruction and stats there? What do you guys usually do? > Thanks a lot for your help, > Nuria > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip _______________________________________________ fieldtrip mailing list fieldtrip at donders.ru.nl http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From kamx at ymail.com Fri Aug 24 06:56:53 2012 From: kamx at ymail.com (Kam X) Date: Thu, 23 Aug 2012 21:56:53 -0700 (PDT) Subject: [FieldTrip] Problem setting up fieldtrip Message-ID: <1345784213.64205.YahooMailClassic@web162305.mail.bf1.yahoo.com> Hi, I hope this is the proper type of questions for this list, if not I apologize.I've just set up fieldtrip by unzipping it, adding the path to the folder to matlab, and running ft_defaults.  Most things seem to work, but if I try to use the command ft_realtime_signalproxy I get: Undefined function 'ft_realtime_signalproxy' for input arguments of type 'struct'. Looking in the fieldtrip folder, I see the .m files, but ft_realtime_signalproxy is not among them.  If I run realtime_signalproxy(cfg), I get: Warning: REALTIME_SIGNALPROXY is only a compatibility wrapper, which will soon be removed.Please instead call FT_REALTIME_SIGNALPROXY. > In realtime_signalproxy at 16 Undefined function 'ft_realtime_signalproxy' for input arguments of type 'struct'. Error in realtime_signalproxy (line 18)[varargout{1:nargout}] = funhandle(varargin{:}); Have I missed a step in setting up fieldtrip, or do I need to download the realtime functions from somewhere? Thanks,Kameron -------------- next part -------------- An HTML attachment was scrubbed... URL: From jm.horschig at donders.ru.nl Fri Aug 24 09:51:06 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Fri, 24 Aug 2012 09:51:06 +0200 Subject: [FieldTrip] Problem setting up fieldtrip In-Reply-To: <1345784213.64205.YahooMailClassic@web162305.mail.bf1.yahoo.com> References: <1345784213.64205.YahooMailClassic@web162305.mail.bf1.yahoo.com> Message-ID: <5037326A.8030209@donders.ru.nl> Dear Kameron, Thanks for reporting this. FieldTrip is modularly organized, with 'realtime' being a separate module/toolbox. Usually, modules should be added automatically when calling ft_defaults. The ft_hastoolbox function should take care of this. However, the realtime folder has been restructured recently so that it has subfolders now, which are (unfortunately) not added by default. It is a(tiny little) bit more tedious now to add this to your path manually: /ft_hastoolbox('realtime/example', 1)/ The same holds for any other subfolders of the realtime module. I'm gonna check with Robert (probably after the BioMag conference) how to resolve this best (if at all). Best, Jörn On 8/24/2012 6:56 AM, Kam X wrote: > Hi, > > I hope this is the proper type of questions for this list, if not I > apologize. > I've just set up fieldtrip by unzipping it, adding the path to the > folder to matlab, and running ft_defaults. Most things seem to work, > but if I try to use the command ft_realtime_signalproxy I get: > > Undefined function 'ft_realtime_signalproxy' for input arguments of > type 'struct'. > > Looking in the fieldtrip folder, I see the .m files, but > ft_realtime_signalproxy is not among them. If I run > realtime_signalproxy(cfg), I get: > > Warning: REALTIME_SIGNALPROXY is only a compatibility wrapper, which > will soon be removed. > Please instead call FT_REALTIME_SIGNALPROXY. > > In realtime_signalproxy at 16 > Undefined function 'ft_realtime_signalproxy' for input arguments of > type 'struct'. > > Error in realtime_signalproxy (line 18) > [varargout{1:nargout}] = funhandle(varargin{:}); > > Have I missed a step in setting up fieldtrip, or do I need to download > the realtime functions from somewhere? > > Thanks, > Kameron > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group FieldTrip Development Team P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From bobrobbrown at googlemail.com Fri Aug 24 10:38:58 2012 From: bobrobbrown at googlemail.com (Robert Brown) Date: Fri, 24 Aug 2012 11:38:58 +0300 Subject: [FieldTrip] PLV on sources Message-ID: Dear Fieldtrippers, I see (from the ft_connectivityanalysis code) that one cannot run phase locking (PLV) analysis (a la Lachaux et al) on source data. My brain still insisted on trying to do so but unfortunately I am not sure if it is clever enough. Hope you can help! It seems that a way to go would be to do LCMV beamforming on the data, get the virtual electrodes, then run freqanalysis with fourier output on these virtual electrodes and then perform PLV on those data. Before I dig myself very deep into this, maybe some experts here could guide me: Would this approach work? Any possible caveats? Are there maybe other, better and more straightforward ways for achieving the PLV on source level? Any comments would be appreciated. Thank you very much for your time! Kind regards, Bob -------------- next part -------------- An HTML attachment was scrubbed... URL: From s.bogels at psy.gla.ac.uk Fri Aug 24 13:25:48 2012 From: s.bogels at psy.gla.ac.uk (Sara =?iso-8859-1?b?QvZnZWxz?=) Date: Fri, 24 Aug 2012 12:25:48 +0100 Subject: [FieldTrip] How to specify two dipoles in cfg.supdip in ft_sourceanalysis? Message-ID: <20120824122548.739016jhmbe93sn0@horde.psy.gla.ac.uk> Hi all, I am using ft_sourceanalysis with specifications of cfg.refdip and cfg.supdip. Until now, I always inserted the same voxel coordinates in supdip as refdip to suppress self-coherence. However, the help suggests that you can also suppress more than one dipole, but it does not specify how to input this in cfg.supdip. I tried a 2 by 3 matrix and a cell structure with two entries (1 x 3 matrices) but that gave (apparently) the same result as suppressing only refdip. As a related, more general question, what about the orientation of the (e.g. reference) dipole? How is that actually determined in this case (since you only have to insert the position of the dipole)? Thank you, Sara ---------------------------------------------------------------- This message was sent using the Web mail system for The University of Glasgow School of Psychology ------------------------------------------------------------------ From narayan.ps at tut.fi Fri Aug 24 14:35:40 2012 From: narayan.ps at tut.fi (Narayan Puthanmadam Subramaniyam) Date: Fri, 24 Aug 2012 12:35:40 +0000 Subject: [FieldTrip] PLV on sources Message-ID: <201208241235.q7OCZhaZ002930@brievenbus.science.ru.nl> hi i was thinking of doing the same thing exactly. Computing plv on the eeg inverse solution. technically it should be possible going by the general level concept of phase locking. but i am not sure so let the experts here speak. Thanks&Regards, NPS Sent from my Nokia phone -----Original Message----- From: Robert Brown Sent: 24:08:2012, 11:38 To: fieldtrip at science.ru.nl Subject: [FieldTrip] PLV on sources Dear Fieldtrippers, I see (from the ft_connectivityanalysis code) that one cannot run phase locking (PLV) analysis (a la Lachaux et al) on source data. My brain still insisted on trying to do so but unfortunately I am not sure if it is clever enough. Hope you can help! It seems that a way to go would be to do LCMV beamforming on the data, get the virtual electrodes, then run freqanalysis with fourier output on these virtual electrodes and then perform PLV on those data. Before I dig myself very deep into this, maybe some experts here could guide me: Would this approach work? Any possible caveats? Are there maybe other, better and more straightforward ways for achieving the PLV on source level? Any comments would be appreciated. Thank you very much for your time! Kind regards, Bob _______________________________________________ fieldtrip mailing list fieldtrip at donders.ru.nl http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From maximilien.chaumon at gmail.com Fri Aug 24 17:13:22 2012 From: maximilien.chaumon at gmail.com (Maximilien Chaumon) Date: Fri, 24 Aug 2012 17:13:22 +0200 Subject: [FieldTrip] [Eeglablist] nested hypothesis testing to decide whether to use one or two dipoles to fit a component In-Reply-To: References: Message-ID: Hi everyone, Just to follow-up on this if anyone wanted to take that approach. I did use the following procedure to decide whether to fit one or two dipoles to a given component. 1) fit one dipole, compute SSE for each component map as the RV * sum of squared inverse weights (sse(1) = EEG.dipfit.model(idip).rv .* sum(EEG.icawinv(:,idip).^2)) 2) fit two dipoles, compute SSE for each component map: sse(2) 3) assume p1 = 6 parameters estimated for 1 dipole, p2 = 9 parameters estimated for 2 dipoles (only three momentum values more, since symmetry imposes position for the two dipoles), n = EEG.nbchan 4) F statistic : F = (sse(1)-sse(2))/(p2-p1)/(sse(2)/(n-p2)) 5) if F > finv(.95,p2-p1,n-p2) decide to go for 2 dipoles. 6) that yielded some solutions where 2 dipoles were selected but were in fact at the same position (x == 0) with opposite momenta. Some others had really just one dipole with large amplitude and the other much smaller. So I discarded solutions where distance between the two dipoles was less than 5mm, or where the momentum angles were opposed, or when the amplitude ratio between the two dipoles was bigger than 5. The final solution is only moderately satisfying. Looking at the maps, the 2 dipole solution is still too often favored over the one dipole solution. As Scott previously mentioned, it could be that the second dipole accounts for error in the head model, and that's all. The F difference in SSE is probably not a good measure of the two-dipolarity of a component. One would probably better just screen all components by eye and spot those that look dipolar using thumbometric visual assessment. Best, Max PS: the script I used: dofit = 1; dorework = 0; doplot = 1; rootdir = '/DATA/'; cd(rootdir) studyfile = fullfile(rootdir,'AP2.study'); if dofit || dorework [STUDY ALLEEG] = pop_loadstudy(studyfile); for i_set = 1:numel(STUDY.datasetinfo) EEG = pop_loadset('filename',ALLEEG(i_set).filename,'filepath',ALLEEG(i_set).filepath);%pop_loadset(GRAB(i_set).name); if isfield(EEG,'peakalphafrequency') EEG.peakalphafreq = EEG.peakalphafrequency; EEG = rmfield(EEG,'peakalphafrequency'); end if ~dorework EEG = pop_dipfit_settings( EEG, 'hdmfile',[ fileparts(which('eeglab')) '/plugins/dipfit2.2/standard_BEM/standard_vol.mat'],'coordformat','MNI','mrifile',[ fileparts(which('eeglab')) '/plugins/dipfit2.2/standard_BEM/standard_mri.mat'],'chanfile',[ fileparts(which('eeglab')) '/plugins/dipfit2.2/standard_BEM/elec/standard_1005.elc'],'coord_transform',[0.67468 -15.6307 2.426 0.081417 0.0024349 -1.5728 1.1744 1.0622 1.1487] ,'chansel',[1:EEG.nbchan] ); EEGonedipole = pop_multifit(EEG, [1:size(EEG.icawinv,2)] ,'threshold',100,'rmout','on','dipoles',1); EEGtwodipoles = pop_multifit(EEG, [1:size(EEG.icawinv,2)] ,'threshold',100,'rmout','on','dipoles',2); else EEGonedipole = EEG; EEGtwodipoles = EEG; EEGonedipole.dipfit.model = EEG.dipfit.onedipole; EEGtwodipoles.dipfit.model = EEG.dipfit.twodipoles; end % H0: one dipole. EEG.dipfit = EEGonedipole.dipfit; % still we store them both. doesn't hurt. EEG.dipfit.onedipole = EEGonedipole.dipfit.model; EEG.dipfit.twodipoles = EEGtwodipoles.dipfit.model; % now for every component, if the rv is much better for the % 2dipoles than for the 1dipole, we'll use the 2dipoles. for i_dip = 1:numel(EEGonedipole.dipfit.model) sse(1) = EEGonedipole.dipfit.model(i_dip).rv .* sum(EEGonedipole.icawinv(:,i_dip).^2); sse(2) = EEGtwodipoles.dipfit.model(i_dip).rv .* sum(EEGtwodipoles.icawinv(:,i_dip).^2); p1 = 6;% six params for 1dipole model p2 = 9;% 9 for 2 dipoles model (assume only symmetry is fixed). n = EEG.nbchan; % N is number of electrodes % F statistic F = (sse(1)-sse(2))/(p2-p1)/(sse(2)/(n-p2)); EEG.dipfit.Fone2two(i_dip) = F; % F decision Fthresh = finv(.999,p2-p1,n-p2); if F > Fthresh % reject H0, take 2 dipoles model % only if x ~= 0 and the two momenta are not just opposite. mom = EEGtwodipoles.dipfit.model(i_dip).momxyz; pos = EEGtwodipoles.dipfit.model(i_dip).posxyz; [t p r] = cart2sph(mom(:,1),mom(:,2),mom(:,3)); t(2) = t(2)+pi;% d = sqrt(sum(diff(pos).^2)); % if the two dipoles are too close and have exactly % opposite angles cond = d < 5 && (diff(abs(t)) < 0.01 || diff(abs(p)) < 0.01); % or if they are exactly on the sagittal plane cond = cond | abs(EEGtwodipoles.dipfit.model(i_dip).posxyz(1)) < 5; % or if the norm of one is much bigger than the other ratio = norm(mom(1,:)) ./ norm(mom(2,:)); ratiothresh = 5; cond = cond | (ratio > ratiothresh || ratio < 1/ratiothresh); % we will still keep the one dipole fit if ~cond% so if not, we take the 2 dipole fit. EEG.dipfit.model(i_dip) = EEGtwodipoles.dipfit.model(i_dip); end end end % reject poor overall fits. EEG.dipfit.model = dipfit_reject(EEG.dipfit.model, 20/100); % reject NaN rv for i_m = 1:numel(EEG.dipfit.model) if isnan(EEG.dipfit.model(i_m).rv) EEG.dipfit.model(i_m).posxyz = []; EEG.dipfit.model(i_m).momxyz = []; EEG.dipfit.model(i_m).rv = 1; end end % removing dipoles outside the head % --------------------------------- rmdip = []; for index = 1:numel(EEGonedipole.dipfit.model) if ~isempty(EEG.dipfit.model(index).posxyz) if any(sqrt(sum(EEG.dipfit.model(index).posxyz.^2,2)) > 85) rmdip = [ rmdip index]; EEG.dipfit.model(index).posxyz = []; EEG.dipfit.model(index).momxyz = []; EEG.dipfit.model(index).rv = 1; end; end; end; if ~isempty(rmdip) fprintf('%d out of cortex dipoles removed (usually artifacts)\n', length(rmdip)); end; pop_saveset(EEG,'savemode','resave'); end end %% if doplot % [STUDY ALLEEG] = pop_loadstudy('AlphaPsyc2.study'); num = [];% number of dipoles that are fit for each component idx = []; topos = {[] [] []}; Fdiff = {[] [] []}; idx = {[] [] []}; dips = {}; for i_e = 1:numel(ALLEEG) for i_comp = 1:numel(ALLEEG(i_e).dipfit.model) num(end+1) = size(ALLEEG(i_e).dipfit.model(i_comp).posxyz,1); topos{num(end)+1}(end+1,:) = ALLEEG(i_e).icawinv(:,i_comp)'; dips{num(end)+1}(size(topos{num(end)+1},1)) = ALLEEG(i_e).dipfit.model(i_comp); Fdiff{num(end)+1}(end+1) = ALLEEG(i_e).dipfit.Fone2two(i_comp); idx{num(end)+1}(end+1,:) = [i_e i_comp]; end end for i_plot = 2:3 figure(6548+i_plot);paste_figpos manytopos ntopos = size(topos{i_plot},1); n = ceil(sqrt(ntopos)); m = n; for i_topo = 1:ntopos subplot(n,m,i_topo) cla topoplot(topos{i_plot}(i_topo,:),ALLEEG(1).chanlocs,'conv','on','electrodes','off'); title({num2str(idx{i_plot}(i_topo,:)) num2str(Fdiff{i_plot}(i_topo))}); drawnow end end end 2012/8/13 Maximilien Chaumon > Hello eeglab & Fieldtrip, > > I'm trying to find out if it would be possible to use a nested hypothesis > testing approach to decide whether to use a one or two dipole model while > estimating components' dipole locations. > The rationale I would like to follow is this: with two dipoles, we will > always obtain a better fit than with one dipole, but the decrease in sum of > squared errors (SSE) should follow a F distribution with k (= > Nparameters_2dipoles - Nparameters_1dipole) degrees of freedom. If the > decrease in SSE is greater than what would be expected under this F > distribution, then we decide that 2 dipoles provide a sufficiently better > fit and decide using them. > > I asked this question to eeglablist and Scott pointed out that it is > difficult/impossible(?) to determine if the second dipole fits actual > interesting data or just noise introduced by the imperfect head model. > Christian then said it'd be worth a shot, and I agree, so here I am again > with two questions, or two confirmations: > > 1) *How many parameters are estimated in ft_dipolefitting.m ?* specially > in the case of 2 dipoles. If I count correctly, we estimate 6 parameters > for one dipole, and, depending on whether the orientation has to be the > same in the 2 dipoles, one (amplitude) or three (amplitude and orientation) > more. > 2) *Can I assimilate the relative residual variance to a SSE?* the > function rv.m does this: rv = sum((d1-d2).^2) ./ sum(d1.^2); > So that seems to be a sum of squared errors divided by the variance of the > original data. So if I multiply the rv by the sum squared component map, I > should get it, right? > > Thanks a lot! > Max > > > 2012/8/11 Christian Kothe > >> I can only speak from my armchair here, but it sounds like it should be >> worth a try - even if you don't get the # of parameters exactly right it >> will probably give you at least some level of complexity control in >> whatever the range of validity is. If it works, it may inspire follow-up >> work (e.g., Bayesian model selection or likelihood ratio tests). >> >> The number of parameters for a 2-dipole model seems to be 3 (xyz) + 4 >> (2x the orientation parameters). Not sure about the momentum, though - you >> might look up the place where the actual function minimization is being >> performed in dipfit (fminunc call?) and see whether these are being >> optimized together with the others. >> >> Christian >> >> On Aug 10, 2012, at 3:29 PM, Scott Makeig wrote: >> >> MAX - Unfortunately, in general using two dipoles rather than one will >> ~always improve the fit. Even if the source is a pure single dipole, a >> second dipole can be used to correct for noise or errors in the forward >> head model. This is less always the case for the constrained >> spatially-symmetric dipole pairs allowed by dipfit(). However, we have not >> thought of an optimal way to decide between using one or (occasionally) two >> dipoles to fit e.g. maps of ICA brain sources. The goal would be to decide >> whether the two-dipole version is fitting noise/forward model error vs >> actual bilateral source generation... >> >> Scott Makeig >> >> On Thu, Aug 9, 2012 at 1:54 AM, Maximilien Chaumon < >> maximilien.chaumon at gmail.com> wrote: >> >>> Hello all, >>> >>> When fitting dipoles to components, we are all sooner or later puzzled >>> by the question whether to use one or two symmetrical dipoles. >>> >>> Would it be correct to put the problems in terms of a nested hypothesis >>> testing? >>> >>> We are fitting a scalp map with one or two parameters and get a residual >>> variance after the fit. >>> Could we not use this residual variance as a measure of the SSE and >>> compute a F statistic to decide whether to use the more complex (with two >>> dipoles) or simpler (with one dipole) of two nested models? >>> If yes, then how would we decide on the number of degrees of freedom? >>> How many free parameters do we have in each case? x,y,z,and two >>> orientations per dipole? how does the imposed symmetry affect that number? >>> Could we really map residual variance to SSE? How many "observations" do we >>> have in that case (see formula below)? >>> >>> I found this formula, for F: >>> F = (SSEF-SSER)/ (kF-kR) / ((1-SSEF)/(N-kF-1)) >>> where >>> SSE is sum of squared errors, >>> k is numbers of parameters, >>> N is number of observations (? what in our case?) >>> F and R indices for full and reduced model respectively (in our case two >>> and one dipole). >>> >>> >>> Thanks a lot for any comment! >>> Best, >>> Max >>> >>> dipfit >>> >>> >>> >>> >>> >>> _______________________________________________ >>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html >>> To unsubscribe, send an empty email to >>> eeglablist-unsubscribe at sccn.ucsd.edu >>> For digest mode, send an email with the subject "set digest mime" to >>> eeglablist-request at sccn.ucsd.edu >>> >> >> >> >> -- >> Scott Makeig, Research Scientist and Director, Swartz Center for >> Computational Neuroscience, Institute for Neural Computation; Prof. of >> Neurosciences (Adj.), University of California San Diego, La Jolla CA >> 92093-0559, http://sccn.ucsd.edu/~scott >> >> _______________________________________________ >> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html >> To unsubscribe, send an empty email to >> eeglablist-unsubscribe at sccn.ucsd.edu >> For digest mode, send an email with the subject "set digest mime" to >> eeglablist-request at sccn.ucsd.edu >> >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Fri Aug 24 19:51:15 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Fri, 24 Aug 2012 19:51:15 +0200 Subject: [FieldTrip] ft_databrowser Message-ID: Hi, Does anybody knows why I can't see whole head topography in ft_databrowser cfg = []; cfg.viewmode ='component'; cfg.layout = 'CTF275.lay'; ft_databrowser(cfg, comp); [image: Inline images 1] Kind regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.jpeg Type: image/jpeg Size: 78272 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ica.JPG Type: image/jpeg Size: 78272 bytes Desc: not available URL: From r.vandermeij at donders.ru.nl Sat Aug 25 11:40:08 2012 From: r.vandermeij at donders.ru.nl (Roemer van der Meij) Date: Sat, 25 Aug 2012 11:40:08 +0200 Subject: [FieldTrip] ft_databrowser In-Reply-To: References: Message-ID: Hi Nenad, This was indeed a bug, and was fixed very recently. If you update to the latest version, it should work fine. Do note the speed-upgrade that I implemented ;), it should be much faster now. Also, just to gather some interest, when you now make a data-selection in the browser, you can right-click to get a context menu with several interesting options, do have a look! :) Best, Roemer On Fri, Aug 24, 2012 at 7:51 PM, Nenad Polomac wrote: > Hi, > > Does anybody knows why I can't see whole head topography in ft_databrowser > > > cfg = []; > cfg.viewmode ='component'; > cfg.layout = 'CTF275.lay'; > ft_databrowser(cfg, comp); > > [image: Inline images 1] > > Kind regards! > > Nenad > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- Roemer van der Meij M.Sc. PhD student Donders Institute for Brain, Cognition and Behaviour Centre for Cognition P.O. Box 9104 6500 HE Nijmegen The Netherlands Tel: +31(0)24 3655932 E-mail: r.vandermeij at donders.ru.nl -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.jpeg Type: image/jpeg Size: 78272 bytes Desc: not available URL: From polomacnenad at gmail.com Sat Aug 25 20:03:09 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Sat, 25 Aug 2012 20:03:09 +0200 Subject: [FieldTrip] fieldtrip Digest, Vol 21, Issue 31 In-Reply-To: References: Message-ID: Dear Roemer, Thank you very much for your suggestion. Now works fine and much faster. Keep up the good work! :) All the best! Nenad > Hi Nenad, > > This was indeed a bug, and was fixed very recently. If you update to the > latest version, it should work fine. Do note the speed-upgrade that I > implemented ;), it should be much faster now. > Also, just to gather some interest, when you now make a data-selection in > the browser, you can right-click to get a context menu with several > interesting options, do have a look! :) > > Best, > Roemer > > > > On Fri, Aug 24, 2012 at 7:51 PM, Nenad Polomac >wrote: > > > Hi, > > > > Does anybody knows why I can't see whole head topography in > ft_databrowser > > > > > > cfg = []; > > cfg.viewmode ='component'; > > cfg.layout = 'CTF275.lay'; > > ft_databrowser(cfg, comp); > > > > [image: Inline images 1] > > > > Kind regards! > > > > Nenad > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > > -- > Roemer van der Meij M.Sc. > PhD student > Donders Institute for Brain, Cognition and Behaviour > Centre for Cognition > P.O. Box 9104 > 6500 HE Nijmegen > The Netherlands > Tel: +31(0)24 3655932 > E-mail: r.vandermeij at donders.ru.nl > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: < > http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20120825/351f7b56/attachment.html > > > -------------- next part -------------- > A non-text attachment was scrubbed... > Name: not available > Type: image/jpeg > Size: 78272 bytes > Desc: not available > URL: < > http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20120825/351f7b56/attachment.jpe > > > > ------------------------------ > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > End of fieldtrip Digest, Vol 21, Issue 31 > ***************************************** > -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Sat Aug 25 20:09:30 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Sat, 25 Aug 2012 20:09:30 +0200 Subject: [FieldTrip] ft_databrowser Message-ID: Dear Roemer, Thank you very much for your suggestion. Now works fine and much faster. Keep up the good work! :) I am sorry but I don't know how to put replay massage in the same thread in the Fieldtrip discussion list . Please give me some clue All the best! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From ayelet.landau at gmail.com Sat Aug 25 21:03:46 2012 From: ayelet.landau at gmail.com (Ayelet Landau) Date: Sat, 25 Aug 2012 21:03:46 +0200 Subject: [FieldTrip] ft_databrowser In-Reply-To: References: Message-ID: Hi all, I thought I would report here also another (perhaps related?) strange behavior that I encountered with databrowser. If you look at the attached image you will immediately note that there is an empty portion after the end of my data. The data plotted is indeed all the data (so, no missing data), but this extra white just appears to varying degrees from trial to trial. any idea why this occurs? thanks! Ayelet [image: Inline image 1] On Sat, Aug 25, 2012 at 8:09 PM, Nenad Polomac wrote: > > Dear Roemer, > > Thank you very much for your suggestion. Now works fine and much faster. > Keep up the good work! :) > > I am sorry but I don't know how to put replay massage in the same thread > in the Fieldtrip discussion list . Please give me some clue > > All the best! > > Nenad > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- Ayelet N. Landau Postdoctoral Scientist, Ernst Strüngmann Institute (ESI) for Neuroscience in Cooperation with Max Planck Society Deutschordenstr. 46, D-60528 Frankfurt Mobile: +49 (0)16 22733 110 Fax: +49 (0)69 96769 555 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.png Type: image/png Size: 44638 bytes Desc: not available URL: From r.vandermeij at donders.ru.nl Sun Aug 26 01:06:28 2012 From: r.vandermeij at donders.ru.nl (Roemer van der Meij) Date: Sun, 26 Aug 2012 01:06:28 +0200 Subject: [FieldTrip] ft_databrowser In-Reply-To: References: Message-ID: Hi Ayelet, This is an intended feature, to make sure each trial is plotted at the same 'zoom-level' (keeping the plotted size of the time-axis the same). We hope this will help in recognizing odd patterns and such. You can also now zoom-in into the time-axis by pressing shift+left or right, or by using the buttons at the lower left corner. Hope this info helps, Best, Roemer On Sat, Aug 25, 2012 at 9:03 PM, Ayelet Landau wrote: > Hi all, > I thought I would report here also another (perhaps related?) strange > behavior that I encountered with databrowser. If you look at the attached > image you will immediately note that there is an empty portion after the > end of my data. The data plotted is indeed all the data (so, no missing > data), but this extra white just appears to varying degrees from trial to > trial. > > any idea why this occurs? > > thanks! > Ayelet > > [image: Inline image 1] > > On Sat, Aug 25, 2012 at 8:09 PM, Nenad Polomac wrote: > >> >> Dear Roemer, >> >> Thank you very much for your suggestion. Now works fine and much faster. >> Keep up the good work! :) >> >> I am sorry but I don't know how to put replay massage in the same thread >> in the Fieldtrip discussion list . Please give me some clue >> >> All the best! >> >> Nenad >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip >> > > > > -- > Ayelet N. Landau > > Postdoctoral Scientist, > Ernst Strüngmann Institute (ESI) for Neuroscience > in Cooperation with Max Planck Society > Deutschordenstr. 46, D-60528 Frankfurt > > Mobile: +49 (0)16 22733 110 > Fax: +49 (0)69 96769 555 > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- Roemer van der Meij M.Sc. PhD student Donders Institute for Brain, Cognition and Behaviour Centre for Cognition P.O. Box 9104 6500 HE Nijmegen The Netherlands Tel: +31(0)24 3655932 E-mail: r.vandermeij at donders.ru.nl -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.png Type: image/png Size: 44638 bytes Desc: not available URL: From karl.doron at gmail.com Sun Aug 26 20:53:53 2012 From: karl.doron at gmail.com (Karl Doron) Date: Sun, 26 Aug 2012 11:53:53 -0700 Subject: [FieldTrip] connectivity/amplitude correlation Message-ID: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> Hello, I'm wondering if there are any publications with the maths behind the functions for ft_connectivity_corr. I would specifically like to confirm (for a pending publication) how the amplitude correlation is computed when the function is given output from ft_freqanalysis which contains a time dimension (relevant cfg parameters pasted below). I've assumed this is occurring in the frequency domain across trials for sensor i and sensor j. cfg.channelcmb = chancmb; cfg.method = 'mtmconvol'; cfg.output = 'fourier'; cfg.keeptrials = 'yes'; cfg.foi = band(1):band(2); cfg.toi = 1.8:0.01:3.0; Thanks for any feedback! Karl Doron, Ph.D. UC, Santa Barbara -------------- next part -------------- An HTML attachment was scrubbed... URL: From stan.vanpelt at fcdonders.ru.nl Mon Aug 27 09:59:02 2012 From: stan.vanpelt at fcdonders.ru.nl (Stan van Pelt) Date: Mon, 27 Aug 2012 09:59:02 +0200 (CEST) Subject: [FieldTrip] ft_databrowser In-Reply-To: Message-ID: <1604964460.83554.1346054342785.JavaMail.root@sculptor.zimbra.ru.nl> Hi Roemer, Interesting features! In that respect, it might also be useful to set the x-axis ticklabels at 'logical'/more intuitive values (e.g., 0.1s apart) instead of 1/xth of the total data length. Best, Stan ----- Oorspronkelijk bericht ----- > Van: "Roemer van der Meij" > Aan: "ayelet landau" , "FieldTrip discussion > list" > Verzonden: Zondag 26 augustus 2012 01:06:28 > Onderwerp: Re: [FieldTrip] ft_databrowser > Hi Ayelet, > This is an intended feature, to make sure each trial is plotted at the > same 'zoom-level' (keeping the plotted size of the time-axis the > same). We hope this will help in recognizing odd patterns and such. > You can also now zoom-in into the time-axis by pressing shift+left or > right, or by using the buttons at the lower left corner. > Hope this info helps, > Best, > Roemer > On Sat, Aug 25, 2012 at 9:03 PM, Ayelet Landau < > ayelet.landau at gmail.com > wrote: > > Hi all, > > I thought I would report here also another (perhaps related?) > > strange > > behavior that I encountered with databrowser. If you look at the > > attached image you will immediately note that there is an empty > > portion after the end of my data. The data plotted is indeed all the > > data (so, no missing data), but this extra white just appears to > > varying degrees from trial to trial. > > any idea why this occurs? > > thanks! > > Ayelet > > Inline image 1 > > On Sat, Aug 25, 2012 at 8:09 PM, Nenad Polomac < > > polomacnenad at gmail.com > wrote: > > > Dear Roemer, > > > Thank you very much for your suggestion. Now works fine and much > > > faster. > > > Keep up the good work! :) > > > I am sorry but I don't know how to put replay massage in the same > > > thread in the Fieldtrip discussion list . Please give me some clue > > > All the best! > > > Nenad > > > _______________________________________________ > > > fieldtrip mailing list > > > fieldtrip at donders.ru.nl > > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > -- > > Ayelet N. Landau > > Postdoctoral Scientist, > > Ernst Strüngmann Institute (ESI) for Neuroscience > > in Cooperation with Max Planck Society > > Deutschordenstr. 46, D-60528 Frankfurt > > Mobile: +49 (0)16 22733 110 > > Fax: +49 (0)69 96769 555 > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- > Roemer van der Meij M.Sc. > PhD student > Donders Institute for Brain, Cognition and Behaviour > Centre for Cognition > P.O. Box 9104 > 6500 HE Nijmegen > The Netherlands > Tel: +31(0)24 3655932 > E-mail: r.vandermeij at donders.ru.nl > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Stan van Pelt Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen Kapittelweg 29, 6525 EN Nijmegen, Netherlands E-mail: stan.vanpelt at donders.ru.nl Website: www.ru.nl/donders/ Phone: (+31) (0)24 36 10981 Fax: (+31) (0)24 36 10989 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.png Type: image/png Size: 44638 bytes Desc: not available URL: From Lilla.Magyari at mpi.nl Mon Aug 27 11:59:53 2012 From: Lilla.Magyari at mpi.nl (Magyari, Lilla) Date: Mon, 27 Aug 2012 11:59:53 +0200 Subject: [FieldTrip] connectivity/amplitude correlation In-Reply-To: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> References: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> Message-ID: <503B4519.5030506@mpi.nl> Dear Karl, There are four papers mentioned in the help of ft_connectivity_corr. I am wondering if any of those could answer your question. Best, Lilla Karl Doron wrote: > Hello, > > I'm wondering if there are any publications with the maths behind the > functions for ft_connectivity_corr. I would specifically like to confirm > (for a pending publication) how the amplitude correlation is computed > when the function is given output from ft_freqanalysis which contains a > time dimension (relevant cfg parameters pasted below). I've assumed this > is occurring in the frequency domain across trials for sensor i and > sensor j. > > cfg.channelcmb = chancmb; > cfg.method = 'mtmconvol'; > cfg.output = 'fourier'; > cfg.keeptrials = 'yes'; > cfg.foi = band(1):band(2); > cfg.toi = 1.8:0.01:3.0; > > > Thanks for any feedback! > > > Karl Doron, Ph.D. > UC, Santa Barbara > > > > ------------------------------------------------------------------------ > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From karl.doron at gmail.com Mon Aug 27 16:16:52 2012 From: karl.doron at gmail.com (Karl Doron) Date: Mon, 27 Aug 2012 07:16:52 -0700 Subject: [FieldTrip] connectivity/amplitude correlation In-Reply-To: <503B4519.5030506@mpi.nl> References: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> <503B4519.5030506@mpi.nl> Message-ID: <0F563C8B-4EA4-43F2-9FF3-93625C9E4A1D@gmail.com> Hi Lilla, There is a reference for Rosenberg et al, 1998 for the partialization method in the code itself but I'm not seeing any in the help section. Thanks, karl On Aug 27, 2012, at 2:59 AM, "Magyari, Lilla" wrote: > Dear Karl, > > There are four papers mentioned in the help of ft_connectivity_corr. I am wondering if any of those could answer your question. > > Best, > Lilla > > Karl Doron wrote: >> Hello, >> I'm wondering if there are any publications with the maths behind the functions for ft_connectivity_corr. I would specifically like to confirm (for a pending publication) how the amplitude correlation is computed when the function is given output from ft_freqanalysis which contains a time dimension (relevant cfg parameters pasted below). I've assumed this is occurring in the frequency domain across trials for sensor i and sensor j. >> cfg.channelcmb = chancmb; >> cfg.method = 'mtmconvol'; >> cfg.output = 'fourier'; >> cfg.keeptrials = 'yes'; >> cfg.foi = band(1):band(2); >> cfg.toi = 1.8:0.01:3.0; >> Thanks for any feedback! >> Karl Doron, Ph.D. >> UC, Santa Barbara >> ------------------------------------------------------------------------ >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From Lilla.Magyari at mpi.nl Mon Aug 27 16:56:37 2012 From: Lilla.Magyari at mpi.nl (Magyari, Lilla) Date: Mon, 27 Aug 2012 16:56:37 +0200 Subject: [FieldTrip] connectivity/amplitude correlation In-Reply-To: <0F563C8B-4EA4-43F2-9FF3-93625C9E4A1D@gmail.com> References: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> <503B4519.5030506@mpi.nl> <0F563C8B-4EA4-43F2-9FF3-93625C9E4A1D@gmail.com> Message-ID: <503B8AA5.6020207@mpi.nl> hi Karl, Maybe your version of FieldTrip is older, and therefore, it does not contain the references yet. You can check the reference documentation of ft_connectivity_corr on the FieldTrip wiki: http://fieldtrip.fcdonders.nl/reference/ft_connectivity_corr (The reference documentation of the functions is always automatically updated according to the latest version of the documentation part (or "help" part) of the script.) The same articles (and more) can be found also under Literature (Method References) http://fieldtrip.fcdonders.nl/references_to_implemented_methods under the title Connectivity Analysis. I have not read the articles myself, so I am not sure if they give the information you need. But I was just wondering if you are aware of the recently available documentation of the function. Best, Lilla Karl Doron wrote: > Hi Lilla, > > There is a reference for Rosenberg et al, 1998 for the partialization method in the code itself but I'm not seeing any in the help section. > > Thanks, > karl > > > On Aug 27, 2012, at 2:59 AM, "Magyari, Lilla" wrote: > >> Dear Karl, >> >> There are four papers mentioned in the help of ft_connectivity_corr. I am wondering if any of those could answer your question. >> >> Best, >> Lilla >> >> Karl Doron wrote: >>> Hello, >>> I'm wondering if there are any publications with the maths behind the functions for ft_connectivity_corr. I would specifically like to confirm (for a pending publication) how the amplitude correlation is computed when the function is given output from ft_freqanalysis which contains a time dimension (relevant cfg parameters pasted below). I've assumed this is occurring in the frequency domain across trials for sensor i and sensor j. >>> cfg.channelcmb = chancmb; >>> cfg.method = 'mtmconvol'; >>> cfg.output = 'fourier'; >>> cfg.keeptrials = 'yes'; >>> cfg.foi = band(1):band(2); >>> cfg.toi = 1.8:0.01:3.0; >>> Thanks for any feedback! >>> Karl Doron, Ph.D. >>> UC, Santa Barbara >>> ------------------------------------------------------------------------ >>> _______________________________________________ >>> fieldtrip mailing list >>> fieldtrip at donders.ru.nl >>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From r.oostenveld at donders.ru.nl Tue Aug 28 10:05:25 2012 From: r.oostenveld at donders.ru.nl (Robert Oostenveld) Date: Tue, 28 Aug 2012 10:05:25 +0200 Subject: [FieldTrip] PLV on sources In-Reply-To: <201208241235.q7OCZhaZ002930@brievenbus.science.ru.nl> References: <201208241235.q7OCZhaZ002930@brievenbus.science.ru.nl> Message-ID: Dear Robert and Nayaran, The data handling in ft_connectivityanalysis is not yet at the level that it can easily deal with all source representations; that is something that we will improve upon in the future. The way we suggest it now is similar to your suggestion and is outlined in detail in the http://fieldtrip.fcdonders.nl/tutorial/connectivity tutorial. Specifically have a look at the second section where as example source (virtual channel) coherence and granger is computed. In that tutorial you would just change cfg.method in plv and you should be rolling. Best regards, Robert O. PS this connectivity tutorial is quite new, so might also be of interest for other people doing connectivity analysis! Op 24 aug. 2012 om 14:35 heeft "Narayan Puthanmadam Subramaniyam" het volgende geschreven: > hi > > i was thinking of doing the same thing exactly. Computing plv on the eeg inverse solution. technically it should be possible going by the general level concept of phase locking. but i am not sure so let the experts here speak. > > Thanks&Regards, > NPS > > Sent from my Nokia phone > -----Original Message----- > From: Robert Brown > Sent: 24:08:2012, 11:38 > To: fieldtrip at science.ru.nl > Subject: [FieldTrip] PLV on sources > > > Dear Fieldtrippers, > > I see (from the ft_connectivityanalysis code) that one cannot run phase > locking (PLV) analysis (a la Lachaux et al) on source data. > > My brain still insisted on trying to do so but unfortunately I am not sure > if it is clever enough. Hope you can help! > > It seems that a way to go would be to do LCMV beamforming on the data, get > the virtual electrodes, then run freqanalysis with fourier output on these > virtual electrodes and then perform PLV on those data. > > Before I dig myself very deep into this, maybe some experts here could > guide me: Would this approach work? Any possible caveats? Are there maybe > other, better and more straightforward ways for achieving the PLV on source > level? > > Any comments would be appreciated. Thank you very much for your time! > > Kind regards, > Bob > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From sherrykhan78 at gmail.com Tue Aug 28 12:34:14 2012 From: sherrykhan78 at gmail.com (Sheraz Khan) Date: Tue, 28 Aug 2012 06:34:14 -0400 Subject: [FieldTrip] PLV on sources In-Reply-To: References: <201208241235.q7OCZhaZ002930@brievenbus.science.ru.nl> Message-ID: Dear Bob, Another possible solution is to do wavelet decomposition in sensor space and then multiply it with your imaging kernel, then do PLV in source space, this will save some computational time, but still you have to deal with EEG/MEG point spread function. Sheraz Khan Martinos Center MGH/MIT/Harvard On Tue, Aug 28, 2012 at 4:05 AM, Robert Oostenveld < r.oostenveld at donders.ru.nl> wrote: > Dear Robert and Nayaran, > > The data handling in ft_connectivityanalysis is not yet at the level that > it can easily deal with all source representations; that is something that > we will improve upon in the future. > > The way we suggest it now is similar to your suggestion and is outlined in > detail in the http://fieldtrip.fcdonders.nl/tutorial/connectivitytutorial. Specifically have a look at the second section where as example > source (virtual channel) coherence and granger is computed. In that > tutorial you would just change cfg.method in plv and you should be rolling. > > Best regards, > Robert O. > > PS this connectivity tutorial is quite new, so might also be of interest > for other people doing connectivity analysis! > > > Op 24 aug. 2012 om 14:35 heeft "Narayan Puthanmadam Subramaniyam" < > narayan.ps at tut.fi> het volgende geschreven: > > > hi > > > > i was thinking of doing the same thing exactly. Computing plv on the eeg > inverse solution. technically it should be possible going by the general > level concept of phase locking. but i am not sure so let the experts here > speak. > > > > Thanks&Regards, > > NPS > > > > Sent from my Nokia phone > > -----Original Message----- > > From: Robert Brown > > Sent: 24:08:2012, 11:38 > > To: fieldtrip at science.ru.nl > > Subject: [FieldTrip] PLV on sources > > > > > > Dear Fieldtrippers, > > > > I see (from the ft_connectivityanalysis code) that one cannot run phase > > locking (PLV) analysis (a la Lachaux et al) on source data. > > > > My brain still insisted on trying to do so but unfortunately I am not > sure > > if it is clever enough. Hope you can help! > > > > It seems that a way to go would be to do LCMV beamforming on the data, > get > > the virtual electrodes, then run freqanalysis with fourier output on > these > > virtual electrodes and then perform PLV on those data. > > > > Before I dig myself very deep into this, maybe some experts here could > > guide me: Would this approach work? Any possible caveats? Are there maybe > > other, better and more straightforward ways for achieving the PLV on > source > > level? > > > > Any comments would be appreciated. Thank you very much for your time! > > > > Kind regards, > > Bob > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -------------- next part -------------- An HTML attachment was scrubbed... URL: From r.oostenveld at donders.ru.nl Thu Aug 30 10:31:17 2012 From: r.oostenveld at donders.ru.nl (Robert Oostenveld) Date: Thu, 30 Aug 2012 10:31:17 +0200 Subject: [FieldTrip] PLV on sources In-Reply-To: References: <201208241235.q7OCZhaZ002930@brievenbus.science.ru.nl> Message-ID: Dear Sheraz, Bob You are completely right; since the fft and the spatial filter multiplication are both linear operations, their order can be interchanged. If nsource>nchan, this will speed things up. For full brain source reconstructions it will be much faster (and require less memory). It would be nice to extend the tutorial with a section that explains precisely this. Best Robert Op 28 aug. 2012 om 12:34 heeft Sheraz Khan het volgende geschreven: > Dear Bob, > > Another possible solution is to do wavelet decomposition in sensor space and then multiply it with your imaging kernel, then do PLV in source space, this will save some computational time, but still you have to deal with EEG/MEG point spread function. > > Sheraz Khan > Martinos Center > MGH/MIT/Harvard > > On Tue, Aug 28, 2012 at 4:05 AM, Robert Oostenveld wrote: > Dear Robert and Nayaran, > > The data handling in ft_connectivityanalysis is not yet at the level that it can easily deal with all source representations; that is something that we will improve upon in the future. > > The way we suggest it now is similar to your suggestion and is outlined in detail in the http://fieldtrip.fcdonders.nl/tutorial/connectivity tutorial. Specifically have a look at the second section where as example source (virtual channel) coherence and granger is computed. In that tutorial you would just change cfg.method in plv and you should be rolling. > > Best regards, > Robert O. > > PS this connectivity tutorial is quite new, so might also be of interest for other people doing connectivity analysis! > > > Op 24 aug. 2012 om 14:35 heeft "Narayan Puthanmadam Subramaniyam" het volgende geschreven: > > > hi > > > > i was thinking of doing the same thing exactly. Computing plv on the eeg inverse solution. technically it should be possible going by the general level concept of phase locking. but i am not sure so let the experts here speak. > > > > Thanks&Regards, > > NPS > > > > Sent from my Nokia phone > > -----Original Message----- > > From: Robert Brown > > Sent: 24:08:2012, 11:38 > > To: fieldtrip at science.ru.nl > > Subject: [FieldTrip] PLV on sources > > > > > > Dear Fieldtrippers, > > > > I see (from the ft_connectivityanalysis code) that one cannot run phase > > locking (PLV) analysis (a la Lachaux et al) on source data. > > > > My brain still insisted on trying to do so but unfortunately I am not sure > > if it is clever enough. Hope you can help! > > > > It seems that a way to go would be to do LCMV beamforming on the data, get > > the virtual electrodes, then run freqanalysis with fourier output on these > > virtual electrodes and then perform PLV on those data. > > > > Before I dig myself very deep into this, maybe some experts here could > > guide me: Would this approach work? Any possible caveats? Are there maybe > > other, better and more straightforward ways for achieving the PLV on source > > level? > > > > Any comments would be appreciated. Thank you very much for your time! > > > > Kind regards, > > Bob > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Thu Aug 30 10:33:28 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Thu, 30 Aug 2012 10:33:28 +0200 Subject: [FieldTrip] neighbours Message-ID: Hi, I am analyzing data obtained on CTF MEG 275 system. However, 4 sensors are broken and data ended up with 271 sensor. So, my question is should I interpolate the missing sensors with FT_CHANNELREPAI *R*? In the case I leave data with 271 sensor, will those missing sensors influence neighbors configuration in the statistic and the source localisation analysis? Thank you in advance! Kind regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 30 21:57:16 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 30 Aug 2012 21:57:16 +0200 Subject: [FieldTrip] connectivity/amplitude correlation In-Reply-To: <0F563C8B-4EA4-43F2-9FF3-93625C9E4A1D@gmail.com> References: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> <503B4519.5030506@mpi.nl> <0F563C8B-4EA4-43F2-9FF3-93625C9E4A1D@gmail.com> Message-ID: <274B2F2C-FE26-4A89-A2EC-84405BB77C2F@donders.ru.nl> Dear Karl, Perhaps I can add one or two things to Lilla's reply. If I understand your question correctly, it pertains specifically to the math behind the amplitude correlation computation, and how this is done in ft_connectivity_corr. It is indeed as you thought, the dimension over which the correlation is computed is indeed the trial dimension. That is, for each channelpair-time-frequency-point the amplitude correlation is computed across trials. A lot of the intelligence already takes place in ft_connectivityanalysis. In your case ft_connectivityanalysis computes the covariance (and variance) across trials between the amplitudes. The only thing that happens in ft_connectivity_corr is the normalisation with the square root of the variance product. Hope this helps, Jan-Mathijs On Aug 27, 2012, at 4:16 PM, Karl Doron wrote: > Hi Lilla, > > There is a reference for Rosenberg et al, 1998 for the partialization method in the code itself but I'm not seeing any in the help section. > > Thanks, > karl > > > On Aug 27, 2012, at 2:59 AM, "Magyari, Lilla" wrote: > >> Dear Karl, >> >> There are four papers mentioned in the help of ft_connectivity_corr. I am wondering if any of those could answer your question. >> >> Best, >> Lilla >> >> Karl Doron wrote: >>> Hello, >>> I'm wondering if there are any publications with the maths behind the functions for ft_connectivity_corr. I would specifically like to confirm (for a pending publication) how the amplitude correlation is computed when the function is given output from ft_freqanalysis which contains a time dimension (relevant cfg parameters pasted below). I've assumed this is occurring in the frequency domain across trials for sensor i and sensor j. >>> cfg.channelcmb = chancmb; >>> cfg.method = 'mtmconvol'; >>> cfg.output = 'fourier'; >>> cfg.keeptrials = 'yes'; >>> cfg.foi = band(1):band(2); >>> cfg.toi = 1.8:0.01:3.0; >>> Thanks for any feedback! >>> Karl Doron, Ph.D. >>> UC, Santa Barbara >>> ------------------------------------------------------------------------ >>> _______________________________________________ >>> fieldtrip mailing list >>> fieldtrip at donders.ru.nl >>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jm.horschig at donders.ru.nl Fri Aug 31 09:50:17 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Fri, 31 Aug 2012 09:50:17 +0200 Subject: [FieldTrip] neighbours In-Reply-To: References: Message-ID: <50406CB9.7090209@donders.ru.nl> Hi Nenad, the question you are asking does not deserve a definite 'yes' or 'no'. In any case, note that when averaging over subjects, missing sensors for one subject will be removed for all other subjects as well. So, if you are talking about different missing sensors per subject, an interpolation will definitely be a wise thing to do. In case you choose for interpolating the missing channels, you can try to check out the new spherical spline interpolation method (see help ft_channelrepair) - theoretically that should result in a quite good reconstruction. But the (standard) nearest neighbour interpolation is fine as well. With respect to your second question: missing sensors will influence the neighbour configuration depending on the method you chose for neighbour selection. For the 'distance' and 'template' method, each neighbouring channel of the missing will have one sensor less (i.e. the missing one). The 'triangulation' method will create triangle neighbour no matter how many channels you have, so it will just ignore the whole in space. In any case, I'd propose that you check out how satisfied you personally are with the neighbours by calling ft_neighbourplot. Just last week, I added the new option that if you use cfg.enableedit='yes', you can interactively change the neighbourhood structure within the neighbourplot. If I were you, I would start from the template method. As an additional piece of information: I am currently working on improving the neighbour templates, and will upload them probably next week. All that advertised, let me also say that having 4 out of 275 sensors missing won't affect your statistics much. Here, it of course depends on your region if interest, research question etc. E.g. if you are interested in posterior alpha power, but miss four temporal channels, there is no reason to worry. So, without knowing what you are looking for and where the missing channels are, I cannot give you a general advise what to do. As a last remark, the sensitiviy of the statistics will of course be influenced by missing sensors, but I doubt that it will matter much. Btw, I am not quite sure how much the spherical spline interpolation will buy you here, you might give it a try. I bet that it will increase your sensitivity more than the nearest neighbour approach, although both might result in rather low increases. But this is just a general gut feeling rather than anything I empirically validated. Best, Jörn On 8/30/2012 10:33 AM, Nenad Polomac wrote: > Hi, > > I am analyzing data obtained on CTF MEG 275 system. However, 4 sensors > are broken and data ended up with 271 sensor. So, my question is > should I interpolate the missing sensors with FT_CHANNELREPAI > _R_? In the > case I leave data with 271 sensor, will those missing sensors > influence neighbors configuration in the statistic and the source > localisation analysis? > > Thank you in advance! > > Kind regards! > > Nenad > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group FieldTrip Development Team P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From akiko.ikkai at gmail.com Fri Aug 31 21:59:00 2012 From: akiko.ikkai at gmail.com (Akiko Ikkai) Date: Fri, 31 Aug 2012 15:59:00 -0400 Subject: [FieldTrip] FT_SCALPCURRENTDENSITY options and elec units Message-ID: Hi, I'm trying to run plv analysis, first acquiring SCD of my data with ft_scalcurrentdensity, and am confused about the difference between cfg.method = 'finite' and 'spline.' Both gave similar results short-range, but 'spline' gives higher plv at long range plv. I'd love it if someone could explain or suggest reading on the difference between these two methods (or is one method preferred over the other in some cases?). Also, I'm not sure whether or not I should convert my elec.pnt values, since the code ft_scalcurrentdensity says that "Note that the skin conductivity, electrode dimensions and the potential all have to be expressed in the same SI units." My elec.pnt are in cm, and I'm using the default conductivity (0.33 S/m). Do I need to do something like elec.pnt = elec.pnt/100 to convert the unit from cm to m? With or without conversion, I got the same results (viewed with ft_topoplotTFR), and I'm a little confused... Thank you in advance for your time! Akiko -- Akiko Ikkai, Ph.D. Postdoctoral Fellow Department of Psychological and Brain Sciences Johns Hopkins University Ames Hall, 3400 N. Charles St. Baltimore, MD 21218 -------------- next part -------------- An HTML attachment was scrubbed... URL: From dohuyn2001 at hotmail.com Wed Aug 1 07:17:37 2012 From: dohuyn2001 at hotmail.com (DoHyun Kim) Date: Wed, 1 Aug 2012 05:17:37 +0000 Subject: [FieldTrip] Question about BCI2000 2D cursor task with Field trip buffer Message-ID: Hi, I'm a college student currently working on the BCI2000 2D cursor task with Field trip buffer. To moves cursors all directions (up, down, left, and right), I followed the description from the wiki that event_up = event; event_up.value = 1; event_down = event; event_down.value = -1; event_null = event; event_null.value = 0; (see http://www.bci2000.org/wiki/index.php/Programming_Tutorial:Working_with_the_FieldTrip_buffer) Currently, my cursor moves right and left with settings as event_right.value=1, event_left.value =-1. While I'm trying to move my cursor on y-axis(since I'm aiming for 2D control), I couldn't figure out how I should write the script for up & down movements. Does anyone know how to do it? Thanks. -------------- next part -------------- An HTML attachment was scrubbed... URL: From vitoria.piai at gmail.com Wed Aug 1 13:22:49 2012 From: vitoria.piai at gmail.com (=?ISO-8859-1?Q?Vit=F3ria_Magalh=E3es_Piai?=) Date: Wed, 01 Aug 2012 13:22:49 +0200 Subject: [FieldTrip] plotting mri ortho with crosshair Message-ID: <50191189.6080809@gmail.com> Hi all, I'm trying to plot an MR image (not necessarily functional data) with the ortho method and with crosshairs but the crosshairs are not there at all. I used to be able to see them before and with interactive mode I was able to go through the coordinates and still seeing where the crosshairs were. What I need in the end is to be able to see on an MRI (without functional data) with crosshairs where exactly a certain coordinate is. Any suggestions? If I remember correctly, I used to plot it using: mri = ft_read_mri(template); cfg.interactive = 'yes'; cfg.crosshair = 'yes'; ft_sourceplot(cfg, mri) Thanx a lot, Vitória -- ** Please consider the environment - do you really need to print? ** From jan.schoffelen at donders.ru.nl Wed Aug 1 13:41:41 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Wed, 1 Aug 2012 13:41:41 +0200 Subject: [FieldTrip] plotting mri ortho with crosshair In-Reply-To: <50191189.6080809@gmail.com> References: <50191189.6080809@gmail.com> Message-ID: Hi Vitoria, Problem confirmed. Could you file this as a bug on bugzilla? We can take it from there. Best, Jan-Mathijs On Aug 1, 2012, at 1:22 PM, Vitória Magalhães Piai wrote: > Hi all, > > I'm trying to plot an MR image (not necessarily functional data) with the ortho method and with crosshairs but the crosshairs are not there at all. > I used to be able to see them before and with interactive mode I was able to go through the coordinates and still seeing where the crosshairs were. > What I need in the end is to be able to see on an MRI (without functional data) with crosshairs where exactly a certain coordinate is. > Any suggestions? > > If I remember correctly, I used to plot it using: > > mri = ft_read_mri(template); > cfg.interactive = 'yes'; > cfg.crosshair = 'yes'; > ft_sourceplot(cfg, mri) > > Thanx a lot, Vitória > > -- > ** Please consider the environment - do you really need to print? ** > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Wed Aug 1 15:00:13 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Wed, 1 Aug 2012 15:00:13 +0200 Subject: [FieldTrip] plotting mri ortho with crosshair In-Reply-To: <50191189.6080809@gmail.com> References: <50191189.6080809@gmail.com> Message-ID: Hi Vitoria, After consulting with Eelke, it seems that the problem is caused by some low-level bug in the graphics handling. More to the point, there seems to be something going on with the openGL rendering in some matlab versions. If you switch to cfg.renderer = 'zbuffer' you should be able to see the cross hair (but you will loose the opacity masking for functional data). For now this is the best we can offer. Cheers, JM On Aug 1, 2012, at 1:22 PM, Vitória Magalhães Piai wrote: > Hi all, > > I'm trying to plot an MR image (not necessarily functional data) with the ortho method and with crosshairs but the crosshairs are not there at all. > I used to be able to see them before and with interactive mode I was able to go through the coordinates and still seeing where the crosshairs were. > What I need in the end is to be able to see on an MRI (without functional data) with crosshairs where exactly a certain coordinate is. > Any suggestions? > > If I remember correctly, I used to plot it using: > > mri = ft_read_mri(template); > cfg.interactive = 'yes'; > cfg.crosshair = 'yes'; > ft_sourceplot(cfg, mri) > > Thanx a lot, Vitória > > -- > ** Please consider the environment - do you really need to print? ** > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Wed Aug 1 17:55:13 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Wed, 1 Aug 2012 17:55:13 +0200 Subject: [FieldTrip] prepare neighbors template file Message-ID: Hi, I am analyzing CTF MEG data with 275 channels. I would like to know where can I find neighbors template file. Or should I create one by myself! I need this file for planar gradient calculation. Thank you in advance! All the best! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From marco.buiatti at gmail.com Wed Aug 1 18:35:53 2012 From: marco.buiatti at gmail.com (Marco Buiatti) Date: Wed, 1 Aug 2012 18:35:53 +0200 Subject: [FieldTrip] interactions between two factors In-Reply-To: References: Message-ID: Thanks Arno and Stephen for your replies. Stephen, the message you cite confirms that what I propose in my first point is correct, thanks. Any statistician willing to answer to my points 2) and 3) is welcome. Best, Marco On 27 July 2012 16:25, Stephen Politzer-Ahles wrote: > Hello Marco, > > For how to test an interaction in a 2x2 design, see > http://mailman.science.ru.nl/pipermail/fieldtrip/2011-January/003447.html > (and some additional information at > http://mailman.science.ru.nl/pipermail/fieldtrip/2010-December/003338.html, > http://mailman.science.ru.nl/pipermail/fieldtrip/2011-September/004244.html, > and Anderson & Braak 2003 in Journal of Statistical Computation & > Simulation). I'm not sure about the other two issues you mentioned, though. > > Best, > Steve > >> Message: 1 >> Date: Thu, 26 Jul 2012 12:36:39 +0200 >> From: Marco Buiatti >> To: fieldtrip at donders.ru.nl >> Subject: Re: [FieldTrip] interactions between two factors >> Message-ID: >> >> >> Content-Type: text/plain; charset=ISO-8859-1 >> >> >> Dear FieldTrippers, >> >> some time ago I have posted the message below concerning how to >> compute the statistical interaction between two factors in an EEG >> study with the FieldTrip cluster-based statistical analysis. Since I >> believe it is a problem of general interest, I was confident I would >> have received some replies, no matter how critics. But I had no reply >> and I am trying to guess why: >> >> - the problem is trivial, I should go back to my statistics books and >> solve it myself; >> - the problem is ill-posed, I should go back to my statistics books >> and reformulate it correctly; >> - the problem is tabou, no one dares commits to a solution because it >> could be a wrong one. >> - the problem is solved: I should read message number #. >> >> Thanks a lot for your feedback, >> >> Best, >> >> Marco >> >> On 25 May 2012 15:58, Marco Buiatti wrote: >> > Dear FieldTrippers, >> > >> > I am analysing an EEG study with 2x4 factors: one varies between 4 >> > parametrically varying levels (1 to 4), the second between two levels. >> > >> > I have three questions concerning the use of Fieldtrip cluster-based >> > non parametric statistical analysis in this case: >> > >> > 1) How to compute the interaction between the two factors. Let's start >> > from the simplest case of a 2x2 design, factors varying between values >> > A1 and A2 for the first factor, B1 and B2 for the second. Please tell >> > me if it is correct to compute the interaction by: >> > - computing the difference diffA=ERP(A1)-ERP(A2) separately in >> > condition B1 and B2, for every subject >> > - performing a within-subjects statistical analysis between diffA in >> > condition B1 and diffA in condition B2 (function >> > statfun_depsamplesT.m). >> > >> > 2) Now consider that factor A varies parametrically between values 1 >> > to 4. For the main effect of this factor, I have used the Fieldtrip >> > function statfun_depsamplesregrT.m and I'm satisfied with it. Is it >> > correct to compute the interaction by >> > - computing the regression >> > regrA=regression(ERP(A1),ERP(A2),ERP(A3),ERP(A4)) (computed as inside >> > function statfun_depsamplesregrT.m) separately in condition B1 and B2, >> > for every subject >> > - performing a within-subjects statistical analysis between regrA in >> > condition B1 and regrA in condition B2 (function >> > statfun_depsamplesT.m)? >> > >> > 3) Since BEFORE looking at the data (this is to prevent Eric's >> > contestation...) I expect a dipolar topography for the regression >> > (data are in average reference), I would like to combine into a joint >> > cluster negative and positive clusters. I have tried by changing >> > statfun_depsamplesregrT.m by just taking the absolute value of the >> > regression, but I get weird results (a huge, non significant cluster). >> > Is it possible that since values are now all positive, I should use a >> > different statistical test at the single bin level? Any other >> > suggestions? >> > >> > Thanks in advance for your help, >> > >> > Marco >> > >> > >> > >> > -- >> > Marco Buiatti, PhD >> > >> > CEA/DSV/I2BM / NeuroSpin >> > INSERM U992 - Cognitive Neuroimaging Unit >> > B?t 145 - Point Courrier 156 >> >> > Gif sur Yvette F-91191 FRANCE >> > Ph: +33(0)169.08.65.21 >> > Fax: +33(0)169.08.79.73 >> > E-mail: marco.buiatti at gmail.com >> > http://www.unicog.org/pm/pmwiki.php/Main/MarcoBuiatti >> > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Marco Buiatti, PhD CEA/DSV/I2BM / NeuroSpin INSERM U992 - Cognitive Neuroimaging Unit Bât 145 - Point Courrier 156 Gif sur Yvette F-91191 FRANCE Ph: +33(0)169.08.65.21 Fax: +33(0)169.08.79.73 E-mail: marco.buiatti at gmail.com http://www.unicog.org/pm/pmwiki.php/Main/MarcoBuiatti *********************************************** From polomacnenad at gmail.com Wed Aug 1 18:41:56 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Wed, 1 Aug 2012 18:41:56 +0200 Subject: [FieldTrip] MEG-anatomical MRI image coregistration Message-ID: Dear Stephen, Thank you very much for your expertise! All the best! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From lihqih at gmail.com Wed Aug 1 19:33:38 2012 From: lihqih at gmail.com (qi li) Date: Wed, 1 Aug 2012 13:33:38 -0400 Subject: [FieldTrip] mne source reconstruction Message-ID: Hi, Is there anyway to map the source space mesh back to voxel space so I can use the ft_sourceplot. After MNE source reconstruction, I have source = time: [1x399 double] pos: [8196x3 double] inside: [1x8196 double] outside: [1x0 double] method: 'average' avg: [1x1 struct] cfg: [1x1 struct] I have 8196 mesh points but how to map it back to the anatomical voxel space? Thanks! Qi From jm.horschig at donders.ru.nl Thu Aug 2 08:49:06 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Thu, 02 Aug 2012 08:49:06 +0200 Subject: [FieldTrip] prepare neighbors template file In-Reply-To: References: Message-ID: <501A22E2.6020505@donders.ru.nl> Dear Nenad, the file can be found in fieldtrip/templates/neighbours It is sufficient to specify cfg.template = 'ctf275_neighb'; or cfg.layout = 'ctf275.lay'; Best, Jörn On 8/1/2012 5:55 PM, Nenad Polomac wrote: > Hi, > I am analyzing CTF MEG data with 275 channels. I would like to know > where can I find neighbors template file. Or should I create one by > myself! I need this file for planar gradient calculation. > > Thank you in advance! > > All the best! > > Nenad > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From jm.horschig at donders.ru.nl Thu Aug 2 08:52:46 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Thu, 02 Aug 2012 08:52:46 +0200 Subject: [FieldTrip] mne source reconstruction In-Reply-To: References: Message-ID: <501A23BE.8040706@donders.ru.nl> Dear Qi, Not quite sure if I understand your request correctly, but you can just rehape: / source.avg.pow = reshape(source.avg.pow, anatomical.dim)/; Then instead of being a vector, it will become a matrix. If that's not the answer you intended to get, please phrase your question differently :) Best, Jörn On 8/1/2012 7:33 PM, qi li wrote: > Hi, > > Is there anyway to map the source space mesh back to voxel space so I > can use the ft_sourceplot. > > After MNE source reconstruction, I have > > source = > > time: [1x399 double] > pos: [8196x3 double] > inside: [1x8196 double] > outside: [1x0 double] > method: 'average' > avg: [1x1 struct] > cfg: [1x1 struct] > > I have 8196 mesh points but how to map it back to the anatomical voxel > space? Thanks! > > Qi > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 2 08:59:04 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 2 Aug 2012 08:59:04 +0200 Subject: [FieldTrip] mne source reconstruction In-Reply-To: References: Message-ID: <5A868F5B-3DD0-4383-BF64-8D53B537A128@donders.ru.nl> Hi Qi, You might want to check the function 'ft_sourceinterpolate' for this. Best, Jan-Mathijs On Aug 1, 2012, at 7:33 PM, qi li wrote: > Hi, > > Is there anyway to map the source space mesh back to voxel space so I > can use the ft_sourceplot. > > After MNE source reconstruction, I have > > source = > > time: [1x399 double] > pos: [8196x3 double] > inside: [1x8196 double] > outside: [1x0 double] > method: 'average' > avg: [1x1 struct] > cfg: [1x1 struct] > > I have 8196 mesh points but how to map it back to the anatomical voxel > space? Thanks! > > Qi > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 2 09:07:22 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 2 Aug 2012 09:07:22 +0200 Subject: [FieldTrip] mne source reconstruction In-Reply-To: <501A23BE.8040706@donders.ru.nl> References: <501A23BE.8040706@donders.ru.nl> Message-ID: <4570E04B-810A-4002-B286-561FDE5C0221@donders.ru.nl> Hi Qi, To follow up on Jörn: If you reconstructed the source activations on a mesh that represents the cortical surface, you first need to interpolate the functional data into a 3D volume, if you want to use ft_sourceplot for visualization. For this you need ft_sourceinterpolate. As an alternative, to visualize the results, you can use ft_sourcemovie. However, this function is under development so it may not give the most visually appealing results. Best, JM On Aug 2, 2012, at 8:52 AM, Jörn M. Horschig wrote: > Dear Qi, > > Not quite sure if I understand your request correctly, but you can just rehape: > source.avg.pow = reshape(source.avg.pow, anatomical.dim); > Then instead of being a vector, it will become a matrix. > > If that's not the answer you intended to get, please phrase your question differently :) > > Best, > Jörn > > On 8/1/2012 7:33 PM, qi li wrote: >> Hi, >> >> Is there anyway to map the source space mesh back to voxel space so I >> can use the ft_sourceplot. >> >> After MNE source reconstruction, I have >> >> source = >> >> time: [1x399 double] >> pos: [8196x3 double] >> inside: [1x8196 double] >> outside: [1x0 double] >> method: 'average' >> avg: [1x1 struct] >> cfg: [1x1 struct] >> >> I have 8196 mesh points but how to map it back to the anatomical voxel >> space? Thanks! >> >> Qi >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > -- > Jörn M. Horschig > PhD Student > Donders Institute for Brain, Cognition and Behaviour > Centre for Cognitive Neuroimaging > Radboud University Nijmegen > Neuronal Oscillations Group > > P.O. Box 9101 > NL-6500 HB Nijmegen > The Netherlands > > Contact: > E-Mail: jm.horschig at donders.ru.nl > Tel: +31-(0)24-36-68493 > Web: http://www.ru.nl/donders > > Visiting address: > Trigon, room 2.30 > Kapittelweg 29 > NL-6525 EN Nijmegen > The Netherlands > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 2 09:16:26 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 2 Aug 2012 09:16:26 +0200 Subject: [FieldTrip] granger causality In-Reply-To: <20120730162938.56630@gmx.net> References: <20120730162938.56630@gmx.net> Message-ID: <17A8F077-820B-48A6-A388-2B2E0B7AB002@donders.ru.nl> Hi Carina, > 1.) I know that I can calculate the appropriate model order and window length with the bsmart toolbox. But from the documentation I somehow do not get the format of data I should use for this calculation. > When I enter the time-frequency data (which I want to use to estimate granger causality) (size: 2 x 2 x freq x time points), it doesn't work. The 'it doesn't work' statement is typically not enough for us to diagnose where the problem lies (at least at which step it goes wrong). At the moment I can only assume that you use an outdated version of FieldTrip, because when I try to do TF-resolved Granger using AR-models it works. It is good that you pasted the script below, but an error message may be informative too. > Any advise? I have a sampling rate of 1000. What sort of time window and maximal model order would make sense do you think? I think that the time window is rather short: I'd rather go for 0.4 or 0.5 to begin with, but this is an empirical question. Also, shifting the windows one time step at a time does not really make sense to me, 0.01 or even 0.05 is much more memory friendly. For the model order: higher one. Some people seem to get nice results when taking a relatively high model order. Of course, for a high model order you also need longer time-windows. > 2)I would like to keep the time domain analyzing my data. What do you think would be an appropriate sliding time window using the 'ft_mvaranalysis' function? And what does it mean exactly? By choosing the model order, I am already determining the maximal time lag between the two functions. Are the values then estimated for the whole time window? No. The time window determines the length of the data segment which is used to fit an AR-model (of order X) of the data at time point Y (as defined in your toi) > 3)What do you think about statistics? Would it makes sense to use the non-parametric cluster approach to shuffle within patients and do a group analysis that way? This depends on your experimental question and you null-hypothesis. Best, Jan-Mathijs > > Thank you so much in advance! > Best, > Carina > > > As a summary,I am doing following steps with field trip: > > > cfg = []; > cfg.dftfilter ='yes' > prep_cond1{subj} = ft_preprocessing(cfg, data); > > > cfg = []; > cfg.order = 5; > cfg.toolbox = 'bsmart'; > cfg.t_ftimwin = 0.05 > cfg.toi = -1:0.001:3.5; > mdata_cond1{subj}= ft_mvaranalysis(cfg, prep_cond1{subj}); > > > cfg = []; > cfg.method = 'mvar'; > cfg.foi = 4:100; > cond1_freq{subj}= ft_freqanalysis_mvar(cfg,mdata_cond1{subj}); > > > cfg = []; > cfg.method = 'granger'; > cond1_granger{subj} = ft_connectivityanalysis(cfg, cond1_freq{subj}); > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 2 09:27:38 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 2 Aug 2012 09:27:38 +0200 Subject: [FieldTrip] Question regarding ft_connectivitysimulation In-Reply-To: <5A1787011651BC42A4D41856DBC2E0603E048720@mbox-f-3.du.deakin.edu.au> References: <5A1787011651BC42A4D41856DBC2E0603E048720@mbox-f-3.du.deakin.edu.au> Message-ID: <6BC78C56-C12D-485A-B916-170A6D0EFCD7@donders.ru.nl> Hi Imali, I am not sure whether I understand completely what you want. I guess you want to simulate time series of 'sources', that have different (more or less well defined) spectral characteristics. As of yet, ft_connectivitysimulation does not support this. You could however call ft_connectivitysimulation twice, (using cfg.method = 'inear_mix') and using different cfg.bpfreq each time. Subsequently, you could sum the resulting time courses: for k = 1:numel(data1.trial) data1.trial{k} = data1.trial{k} + data2.trial{k}; end where data1 and data2 are the output to ft_connectivitysimulation. Best, Jan-Mathijs On Jul 19, 2012, at 6:44 AM, IMALI THANUJA HETTIARACHCHI wrote: > Dear fieldTrippers, > > I am trying to generate some simulated signals from a known connectivity structure with ft_connectivitysimulation, and secondly to assign these signals to known dipole locations when using in ft_dipolesimulation. > > Is there any way I can assign a frequency to the signals in ft_connectivityanalysis, so that I can assign different known frequency signals for dipoles? > > Any help is really appreciated. > > Thank you very much. > > Kind regards > Imali > > Imali Thanuja Hettiarachchi > PhD Candidate > Centre for Intelligent Systems research > Deakin University, Geelong 3217, Australia. > Email: ith at deakin.edu.au > www.deakin.edu.au/cisr > > > > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 2 09:31:03 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 2 Aug 2012 09:31:03 +0200 Subject: [FieldTrip] About co-register the individual sudbject MRI and sensor locations In-Reply-To: <5D2CC5653262C14188174E20065A4C4EC8D457@hscmbx6.uthsc.tennessee.edu> References: <5D2CC5653262C14188174E20065A4C4EC8D457@hscmbx6.uthsc.tennessee.edu> Message-ID: <3CC4394E-B87C-46A8-80C8-B002D2FF79C5@donders.ru.nl> Hi Xiaoxiao, It looks as if your mri (from which you obtained the vol-variable) and the data in the hs_file are not expressed in the same coordinate system. The geometrical objects that you need to combine in order to compute leadfields should be defined in the same coordinate system. My guess is that you will first need to apply ft_volumerealign to your mri_nom variable, to get it back into the Bti-based coordinate system. Best, Jan-Mathijs On Jul 21, 2012, at 1:28 AM, Bai, XiaoXiao wrote: > Dear Sir/Madam, > > I am a new fieldtrip user and want to apply the beamformer method in Fieldtrip for the MEG data from 4D/BTI system. Now I can get a individual subject MRI in analysis/spm format from CURRY software. > > This is my code for create volume head model for the beamformer, > > mri_nom = ft_read_mri(mrifile); > > cfg = []; > [segmentedmri] = ft_volumesegment(cfg, mri_nom); > > cfg = []; > cfg.method = 'singleshell'; > vol = ft_prepare_headmodel(cfg, segmentedmri); > > The vol and sensor locations were displayed and attached with here. > > How can I co-register the vol with sensors based on headshpe file (hs_file) for computing lead field for next step? > > Thanks a lot. > > Best regards, > > Xiaoxiao > Division of Clinical Neurosciences, Department of Pediatrics > University of Tennessee Health Science Center, College of Medicine & > Neuroscience Institute, LeBonheur Children's Hospital > 777 Washington Avenue, P335 > Memphis, TN 38105, USA > Phone: 901-287-4612 > Fax: 901-287-5325 > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Thu Aug 2 14:41:32 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Thu, 2 Aug 2012 14:41:32 +0200 Subject: [FieldTrip] prepare neighbors template file Message-ID: Dear Jörn, Thank you very much for fast answer. I have one more problem concerning this thing. When I do ft_timelockgrandaverage I got this warning: "discarding gradiometer information because it cannot be averaged". So, I tried different ways to calculate ft_megrealign. And I am not sure at which point I should do this function. Before trial averaging in single subject or after? Kind regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From lihqih at gmail.com Thu Aug 2 17:00:50 2012 From: lihqih at gmail.com (qi li) Date: Thu, 2 Aug 2012 11:00:50 -0400 Subject: [FieldTrip] mne source reconstruction In-Reply-To: <4570E04B-810A-4002-B286-561FDE5C0221@donders.ru.nl> References: <501A23BE.8040706@donders.ru.nl> <4570E04B-810A-4002-B286-561FDE5C0221@donders.ru.nl> Message-ID: Jan and Jörn: Thanks a lot for your answers! You did understand my questions pretty well. I actually tried to use ft_sourceinterpolate to find the voxel index without success. The error message is like 'Reference to non-existent field 'sphereradius'. Error in ft_sourceinterpolate (line 203) interpmat = interp_ungridded(functional.pos, warp_apply(anatomical.transform, [X(:) Y(:) Z(:)]), ...' On Thu, Aug 2, 2012 at 3:07 AM, jan-mathijs schoffelen wrote: > Hi Qi, > > To follow up on Jörn: > > If you reconstructed the source activations on a mesh that represents the > cortical surface, you first need to interpolate the functional data into a > 3D volume, if you want to use ft_sourceplot for visualization. For this you > need ft_sourceinterpolate. > As an alternative, to visualize the results, you can use ft_sourcemovie. > However, this function is under development so it may not give the most > visually appealing results. > > Best, > > JM > > > On Aug 2, 2012, at 8:52 AM, Jörn M. Horschig wrote: > > Dear Qi, > > Not quite sure if I understand your request correctly, but you can just > rehape: > source.avg.pow = reshape(source.avg.pow, anatomical.dim); > Then instead of being a vector, it will become a matrix. > > If that's not the answer you intended to get, please phrase your question > differently :) > > Best, > Jörn > > On 8/1/2012 7:33 PM, qi li wrote: > > Hi, > > Is there anyway to map the source space mesh back to voxel space so I > can use the ft_sourceplot. > > After MNE source reconstruction, I have > > source = > > time: [1x399 double] > pos: [8196x3 double] > inside: [1x8196 double] > outside: [1x0 double] > method: 'average' > avg: [1x1 struct] > cfg: [1x1 struct] > > I have 8196 mesh points but how to map it back to the anatomical voxel > space? Thanks! > > Qi > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > -- > Jörn M. Horschig > PhD Student > Donders Institute for Brain, Cognition and Behaviour > Centre for Cognitive Neuroimaging > Radboud University Nijmegen > Neuronal Oscillations Group > > P.O. Box 9101 > NL-6500 HB Nijmegen > The Netherlands > > Contact: > E-Mail: jm.horschig at donders.ru.nl > Tel: +31-(0)24-36-68493 > Web: http://www.ru.nl/donders > > Visiting address: > Trigon, room 2.30 > Kapittelweg 29 > NL-6525 EN Nijmegen > The Netherlands > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > Jan-Mathijs Schoffelen, MD PhD > > Donders Institute for Brain, Cognition and Behaviour, > Centre for Cognitive Neuroimaging, > Radboud University Nijmegen, The Netherlands > > Max Planck Institute for Psycholinguistics, > Nijmegen, The Netherlands > > J.Schoffelen at donders.ru.nl > Telephone: +31-24-3614793 > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From nenga at gmx.net Thu Aug 2 18:51:28 2012 From: nenga at gmx.net (Carina Oehrn) Date: Thu, 02 Aug 2012 18:51:28 +0200 Subject: [FieldTrip] granger causality: model order In-Reply-To: <17A8F077-820B-48A6-A388-2B2E0B7AB002@donders.ru.nl> References: <20120730162938.56630@gmx.net> <17A8F077-820B-48A6-A388-2B2E0B7AB002@donders.ru.nl> Message-ID: <20120802165128.149370@gmx.net> Hey Jan-Mathijs, thank you very much for your response and your advise on the length of time windows! TF-resolved Granger works fine with fieldtrip now. I still have my questions regarding the choice of model order and yes, sorry, my descriptions were not detailed enough. 1) what I meant was that I have problems using the aic_test function of the bsmart toolbox to calculate the most appropriate model order. When I am trying to use maximal model orders above 15 (I have 3000 data points and I tried various different time windows), I always get the error message: ??? Error using ==> aic_test at 47 Cannot compute AIC with 'opssaic'! But I can not find the reason. Others using this function seemed to have the same problem. However, as my data was sampled at a rate of 1000 Hz and as I do not downsample in my analysis, I would like to use orders above 15 (which to my knowledge most people use- who however mostly have smaller sampling rates) to include a time period longer than 15 ms. 2) What range of model orders were you talking about when stating: > For the model order: higher one. Some people seem to get nice results when taking a relatively high model order. Or does anybody else have some experience with higher model orders? How far up do you go to include a sufficient time interval but still avoid overparameterization? Thank you very much for your help! All the best, Carina -------- Original-Nachricht -------- > Datum: Thu, 2 Aug 2012 09:16:26 +0200 > Von: jan-mathijs schoffelen > An: FieldTrip discussion list > Betreff: Re: [FieldTrip] granger causality > Hi Carina, > > > 1.) I know that I can calculate the appropriate model order and window > length with the bsmart toolbox. But from the documentation I somehow do not > get the format of data I should use for this calculation. > > When I enter the time-frequency data (which I want to use to estimate > granger causality) (size: 2 x 2 x freq x time points), it doesn't work. > > The 'it doesn't work' statement is typically not enough for us to diagnose > where the problem lies (at least at which step it goes wrong). At the > moment I can only assume that you use an outdated version of FieldTrip, because > when I try to do TF-resolved Granger using AR-models it works. It is good > that you pasted the script below, but an error message may be informative > too. > > > Any advise? I have a sampling rate of 1000. What sort of time window and > maximal model order would make sense do you think? > > I think that the time window is rather short: I'd rather go for 0.4 or 0.5 > to begin with, but this is an empirical question. Also, shifting the > windows one time step at a time does not really make sense to me, 0.01 or even > 0.05 is much more memory friendly. For the model order: higher one. Some > people seem to get nice results when taking a relatively high model order. Of > course, for a high model order you also need longer time-windows. > > > 2)I would like to keep the time domain analyzing my data. What do you > think would be an appropriate sliding time window using the 'ft_mvaranalysis' > function? And what does it mean exactly? By choosing the model order, I am > already determining the maximal time lag between the two functions. Are > the values then estimated for the whole time window? > > No. The time window determines the length of the data segment which is > used to fit an AR-model (of order X) of the data at time point Y (as defined > in your toi) > > > 3)What do you think about statistics? Would it makes sense to use the > non-parametric cluster approach to shuffle within patients and do a group > analysis that way? > > This depends on your experimental question and you null-hypothesis. > > Best, > Jan-Mathijs > > > > > Thank you so much in advance! > > Best, > > Carina > > > > > > As a summary,I am doing following steps with field trip: > > > > > > cfg = []; > > cfg.dftfilter ='yes' > > prep_cond1{subj} = ft_preprocessing(cfg, data); > > > > > > cfg = []; > > cfg.order = 5; > > cfg.toolbox = 'bsmart'; > > cfg.t_ftimwin = 0.05 > > cfg.toi = -1:0.001:3.5; > > mdata_cond1{subj}= ft_mvaranalysis(cfg, prep_cond1{subj}); > > > > > > cfg = []; > > cfg.method = 'mvar'; > > cfg.foi = 4:100; > > cond1_freq{subj}= ft_freqanalysis_mvar(cfg,mdata_cond1{subj}); > > > > > > cfg = []; > > cfg.method = 'granger'; > > cond1_granger{subj} = ft_connectivityanalysis(cfg, cond1_freq{subj}); > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > Jan-Mathijs Schoffelen, MD PhD > > Donders Institute for Brain, Cognition and Behaviour, > Centre for Cognitive Neuroimaging, > Radboud University Nijmegen, The Netherlands > > Max Planck Institute for Psycholinguistics, > Nijmegen, The Netherlands > > J.Schoffelen at donders.ru.nl > Telephone: +31-24-3614793 > From fredericroux at hotmail.de Fri Aug 3 11:38:31 2012 From: fredericroux at hotmail.de (Frederic Roux) Date: Fri, 3 Aug 2012 11:38:31 +0200 Subject: [FieldTrip] PhD position Message-ID: PhD position Beginning in October 2012, the Brain Imaging Center, Department of Neurology, Goethe-University, offers a PhD position (3years, 65% of standard weekly hours) in the project: "Spectro-temporal dynamics of sensorimotor interactions underlying lateralization of speech production" funded by the German Research Council. The successful applicant will study speech production using magnetoencephalography, electro- corticography, and computational modelling. Our interdisciplinary Emmy Noether group focuses on the question how network dynamics contribute to the generation of complex behavior and offers - besides expertise in functional neuroimaging and Cognitive Neuroscience - a creative and cooperative work environment. We are seeking for an enthusiastic thesis student with interest in the Neurosciences who has advanced skills in Matlab programming and experience in signal processing and/or brain data analysis. He/she should have a background in Natural or Applied Science or Psychology. Electronic applications (no original documents) are requested until 2012/08/31 and should be addressed to: Dr. Christian Kell, Cognitive Neuroscience Group, Brain Imaging Center and Department of Neurology, Goethe University, Schleusenweg 2-16, 60528 Frankfurt, Germany c.kell at em.uni-frankfurt.de www.brainclocks.com Pay group E13 TV-GU. Teaching is not mandatory. Goethe University is an equal opportunity employer. Applications from female candidates are encouraged. Where qualifications are equal, preference will be given to people with disabilities. -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Fri Aug 3 15:36:53 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Fri, 3 Aug 2012 15:36:53 +0200 Subject: [FieldTrip] planar gradient literature Message-ID: Hi, I would like to ask for some literature concerning planar gradient computation. I am new in MEG field and I need some more information on theoretical background of such a calculation. The stuff available in "Event related averaging and planar gradient" tutorial is useful but not very detailed. So, could you please direct me to some reference. Thank you in advance! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From Lilla.Magyari at mpi.nl Fri Aug 3 16:10:31 2012 From: Lilla.Magyari at mpi.nl (Lilla.Magyari at mpi.nl) Date: Fri, 3 Aug 2012 16:10:31 +0200 (CEST) Subject: [FieldTrip] planar gradient literature In-Reply-To: References: Message-ID: <50094.131.174.45.70.1344003031.squirrel@131.174.45.70> Hi Nenad, I usually refer to this paper when I use the planar gradient computation: Bastiaansen, M. C. M., & Knösche, T. R. (2000). MEG tangential derivative mapping applied to Event-Related Desynchronization (ERD) research. Clinical Neurophysiology, 111, 1300-1305. Best, Lilla > Hi, > > I would like to ask for some literature concerning planar gradient > computation. I am new in MEG field and I need some more information on > theoretical background of such a calculation. The stuff available in > "Event > related averaging and planar gradient" tutorial is useful but not > very detailed. So, could you please direct me to some reference. > > Thank you in advance! > > Nenad > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From polomacnenad at gmail.com Fri Aug 3 18:45:48 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Fri, 3 Aug 2012 18:45:48 +0200 Subject: [FieldTrip] planar gradient literature Message-ID: Thank you Lilla! :) All the best! -------------- next part -------------- An HTML attachment was scrubbed... URL: From juan.garciaprieto at ctb.upm.es Sun Aug 5 20:41:35 2012 From: juan.garciaprieto at ctb.upm.es (Juan Garcia-Prieto) Date: Sun, 5 Aug 2012 20:41:35 +0200 Subject: [FieldTrip] anonymize registers Message-ID: Hi everybody, First time mailing this list. great pleasure. This is Juan, from madrid's MEG lab. I've coded a little matlab-FT script which anonymizes fif files (elekta neuromag). Brief story is that when sending a .fif file somewhere else for post-processing, it might be wise to eliminate any sensible information in terms subjects name, id, etc. because of data privacy concerns. When dealing with this issue, Elekta very gently supplied a script wrapper based on regular expressions, to be run under linux. So my effort will only be another way of solving the problem. The script uses "fiff_read_tag_info" "fiff_read_tag" "fiff_make_dir_tree" and "fiff_dir_tree_find" functions (part of FT) by Matti Hamalainen. My question, is: Will this function be useful to some others? Specially taking into account I have ONLY worked with neuromag .fif files. thank you for your work. I've learned a LOT, only reading FT scripts. Best regards, Juan, -------------- next part -------------- An HTML attachment was scrubbed... URL: From kashyapmanik at gmail.com Tue Aug 7 08:38:07 2012 From: kashyapmanik at gmail.com (Manik Gupta) Date: Tue, 7 Aug 2012 07:38:07 +0100 Subject: [FieldTrip] Help needed for reading a time series data Message-ID: Dear All I am new to fieldtrip and want to use to preprocess some air pollution data using fieldtrip. The data is sampled at a frequency of 1 sample/second and currently is in a text file format. Can someone please help me how to process the text file into fieldtrip raw data structure. Thanks a lot, Manik. -------------- next part -------------- An HTML attachment was scrubbed... URL: From stan.vanpelt at fcdonders.ru.nl Tue Aug 7 09:43:38 2012 From: stan.vanpelt at fcdonders.ru.nl (Stan van Pelt) Date: Tue, 7 Aug 2012 09:43:38 +0200 (CEST) Subject: [FieldTrip] Help needed for reading a time series data In-Reply-To: Message-ID: <29876546.58683.1344325418384.JavaMail.root@sculptor.zimbra.ru.nl> Dear Manik, Great that you're trying to apply FieldTrip to non-neuroscientific data. As you probably know, FT is developed for the analysis of electrophysiological data, but that does not mean it cannot be useful to you. W.r.t. your data format, i think it is easiest if you read in the data into Matlab using a function like readtxt.m, and than put these data into a FT-type structure, and run your actual analyses. At the minimum, this structure should look like this: data = label: {187x1 cell} % channel labels (e.g. 'air pollution site 1','air pollution site 2', etc) trial: {1x266 cell} % data (Nchans*Nsamples) for each trial (if you have only one long data set, it means you have only 1 trial; if you have multiple 'recording sessions' from the same site, you would have more) time: {1x266 cell} % time axis for each trial fsample: 300 % sampling frequency It would be great if you would be willing to share your results when they are published and add a FieldTrip reference (http://www.hindawi.com/journals/cin/2011/156869/). Good luck! Stan -- Stan van Pelt, PhD Ernst Strüngmann Institute (ESI) for Neuroscience in Cooperation with Max Planck Society Deutschordenstr. 46 60528 Frankfurt, Germany Website: www.esi-frankfurt.de E-mail: stan.vanpelt at esi-frankfurt.de Tel: +49 (0)69 96769 519 Fax: +49 (0)69 96769 555 ----- Oorspronkelijk bericht ----- > Van: "Manik Gupta" > Aan: fieldtrip at science.ru.nl > Verzonden: Dinsdag 7 augustus 2012 08:38:07 > Onderwerp: [FieldTrip] Help needed for reading a time series data > Dear All > I am new to fieldtrip and want to use to preprocess some air pollution > data using fieldtrip. > The data is sampled at a frequency of 1 sample/second and currently is > in a text file format. > Can someone please help me how to process the text file into fieldtrip > raw data structure. > Thanks a lot, > Manik. > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Stan van Pelt Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen Kapittelweg 29, 6525 EN Nijmegen, Netherlands E-mail: stan.vanpelt at donders.ru.nl Website: www.ru.nl/donders/ Phone: (+31) (0)24 36 10981 Fax: (+31) (0)24 36 10989 -------------- next part -------------- An HTML attachment was scrubbed... URL: From kashyapmanik at gmail.com Tue Aug 7 09:51:31 2012 From: kashyapmanik at gmail.com (Manik Gupta) Date: Tue, 7 Aug 2012 08:51:31 +0100 Subject: [FieldTrip] Help needed for reading a time series data In-Reply-To: <29876546.58683.1344325418384.JavaMail.root@sculptor.zimbra.ru.nl> References: <29876546.58683.1344325418384.JavaMail.root@sculptor.zimbra.ru.nl> Message-ID: Dear Stan, Thanks a lot for the reply. I actually want to do a transfer entropy analysis using Trentool and for that my data needs to be in the fieldtrip raw format. I figured out how to use the data now and will get back to you in case of more queries. Just a quick check, would you be able to provide any assistance on Trentool as well. and I would surely publish the results if they are encouraging enough!! :-) Thanks once again, Manik On Tue, Aug 7, 2012 at 8:43 AM, Stan van Pelt wrote: > Dear Manik, > > Great that you're trying to apply FieldTrip to non-neuroscientific data. > As you probably know, FT is developed for the analysis of > electrophysiological data, but that does not mean it cannot be useful to > you. > W.r.t. your data format, i think it is easiest if you read in the data > into Matlab using a function like readtxt.m, and than put these data into a > FT-type structure, and run your actual analyses. At the minimum, this > structure should look like this: > > data = > > label: {187x1 cell} % channel labels (e.g. 'air pollution site 1','air pollution site 2', etc) > trial: {1x266 cell} % data (Nchans*Nsamples) for each trial (if you have only one long data set, it means you have only 1 trial; if you have multiple 'recording sessions' from the same site, you would have more) > time: {1x266 cell} % time axis for each trial > fsample: 300 % sampling frequency > > It would be great if you would be willing to share your results when they > are published and add a FieldTrip reference ( > http://www.hindawi.com/journals/cin/2011/156869/). > > Good luck! > > Stan > > -- > Stan van Pelt, PhD > > Ernst Strüngmann Institute (ESI) > for Neuroscience in Cooperation with Max Planck Society > Deutschordenstr. 46 > 60528 Frankfurt, Germany > Website: www.esi-frankfurt.de > E-mail: stan.vanpelt at esi-frankfurt.de > Tel: +49 (0)69 96769 519 > Fax: +49 (0)69 96769 555 > > > ------------------------------ > > *Van: *"Manik Gupta" > *Aan: *fieldtrip at science.ru.nl > *Verzonden: *Dinsdag 7 augustus 2012 08:38:07 > *Onderwerp: *[FieldTrip] Help needed for reading a time series data > > > Dear All > > I am new to fieldtrip and want to use to preprocess some air pollution > data using fieldtrip. > The data is sampled at a frequency of 1 sample/second and currently is in > a text file format. > Can someone please help me how to process the text file into fieldtrip raw > data structure. > > Thanks a lot, > Manik. > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > -- > Stan van Pelt > > Donders Institute for Brain, Cognition and Behaviour, Radboud University > Nijmegen > Kapittelweg 29, 6525 EN Nijmegen, Netherlands > E-mail: stan.vanpelt at donders.ru.nl > Website: www.ru.nl/donders/ > Phone: (+31) (0)24 36 10981 > Fax: (+31) (0)24 36 10989 > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -------------- next part -------------- An HTML attachment was scrubbed... URL: From johanna.zumer at donders.ru.nl Tue Aug 7 14:12:20 2012 From: johanna.zumer at donders.ru.nl (Johanna Zumer) Date: Tue, 7 Aug 2012 14:12:20 +0200 Subject: [FieldTrip] mne source reconstruction In-Reply-To: References: <501A23BE.8040706@donders.ru.nl> <4570E04B-810A-4002-B286-561FDE5C0221@donders.ru.nl> Message-ID: Dear Qi, Are you using the most current FieldTrip version? That line is 217 in the current version. But cfg.sphereradius should be set a few lines previous to that line, so I'm not sure how it is a non-existent field. What are the cfg options you use to call ft_sourceinterpolate? Best, Johanna 2012/8/2 qi li > Jan and Jörn: > > Thanks a lot for your answers! You did understand my questions pretty > well. I actually tried to use ft_sourceinterpolate to find the voxel > index without success. The error message is like > > 'Reference to non-existent field 'sphereradius'. > > Error in ft_sourceinterpolate (line 203) > interpmat = interp_ungridded(functional.pos, > warp_apply(anatomical.transform, [X(:) Y(:) Z(:)]), ...' > > On Thu, Aug 2, 2012 at 3:07 AM, jan-mathijs schoffelen > wrote: > > Hi Qi, > > > > To follow up on Jörn: > > > > If you reconstructed the source activations on a mesh that represents the > > cortical surface, you first need to interpolate the functional data into > a > > 3D volume, if you want to use ft_sourceplot for visualization. For this > you > > need ft_sourceinterpolate. > > As an alternative, to visualize the results, you can use ft_sourcemovie. > > However, this function is under development so it may not give the most > > visually appealing results. > > > > Best, > > > > JM > > > > > > On Aug 2, 2012, at 8:52 AM, Jörn M. Horschig wrote: > > > > Dear Qi, > > > > Not quite sure if I understand your request correctly, but you can just > > rehape: > > source.avg.pow = reshape(source.avg.pow, anatomical.dim); > > Then instead of being a vector, it will become a matrix. > > > > If that's not the answer you intended to get, please phrase your question > > differently :) > > > > Best, > > Jörn > > > > On 8/1/2012 7:33 PM, qi li wrote: > > > > Hi, > > > > Is there anyway to map the source space mesh back to voxel space so I > > can use the ft_sourceplot. > > > > After MNE source reconstruction, I have > > > > source = > > > > time: [1x399 double] > > pos: [8196x3 double] > > inside: [1x8196 double] > > outside: [1x0 double] > > method: 'average' > > avg: [1x1 struct] > > cfg: [1x1 struct] > > > > I have 8196 mesh points but how to map it back to the anatomical voxel > > space? Thanks! > > > > Qi > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > > > > -- > > Jörn M. Horschig > > PhD Student > > Donders Institute for Brain, Cognition and Behaviour > > Centre for Cognitive Neuroimaging > > Radboud University Nijmegen > > Neuronal Oscillations Group > > > > P.O. Box 9101 > > NL-6500 HB Nijmegen > > The Netherlands > > > > Contact: > > E-Mail: jm.horschig at donders.ru.nl > > Tel: +31-(0)24-36-68493 > > Web: http://www.ru.nl/donders > > > > Visiting address: > > Trigon, room 2.30 > > Kapittelweg 29 > > NL-6525 EN Nijmegen > > The Netherlands > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > > Jan-Mathijs Schoffelen, MD PhD > > > > Donders Institute for Brain, Cognition and Behaviour, > > Centre for Cognitive Neuroimaging, > > Radboud University Nijmegen, The Netherlands > > > > Max Planck Institute for Psycholinguistics, > > Nijmegen, The Netherlands > > > > J.Schoffelen at donders.ru.nl > > Telephone: +31-24-3614793 > > > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -------------- next part -------------- An HTML attachment was scrubbed... URL: From cs at imm.dtu.dk Tue Aug 7 14:40:46 2012 From: cs at imm.dtu.dk (Carsten Stahlhut) Date: Tue, 7 Aug 2012 14:40:46 +0200 Subject: [FieldTrip] Release of the SmartphoneBrainScanner2 Message-ID: [Apologies for cross-posting] Dear Fieldtrip list We are very happy to announce the release of the SmartphoneBrainScanner2 platform: http://code.google.com/p/smartphonebrainscanner2/ SmartphoneBrainScanner2 is a framework for building cross-platform real-time EEG applications. Originally developed at the Technical University of Denmark for collecting and analyzing signals from Emotiv EPOC headset, its extensible architecture allows working with various EEG systems and multiple platforms. *Cross Platform* SmartphoneBrainScanner2 is written in Qt, a C++ framework offering the power of the native development and unified support for multiple platforms. Plus the UI can be created in QML, high-level declarative UI framework. SBS2 can be compiled for every platform supporting Qt 4, including Linux, OSX, Windows, Android, Maemo 5, MeeGo. Although not yet attempted, it should also work on iOS and BlackBerry OS. *Advanced EEG* SmartphoneBrainScanner2 contains state-of-the-art techniques for working with multi-channel EEG signal in real-time, most notably source reconstruction methods with online adaptation to the noise level. Current implemented source reconstruction approaches cover the minimum-norm and low resolution tomography (LORETA) methods formulated in a Bayesian framework using a expectation-maximization scheme for hyperparameter estimation. The SBS2 source reconstruction is realized using a pre-build forward model connecting the cortical surface with the electrodes at the scalp. The current forward model provided with the software is a 3-spheres model obtained from the Matlab toolbox SPM8 using coarse spatial resolution and with sensor positions in accordance with the Emotiv EPOC system Besides, source reconstruction methods, additional machine learning methods such as independent component analysis (ICA), common spatial patterns (CSP), and Bayesian classifiers are continuously added. *New Approach* Real-time EEG doesn't have to happen in the lab! Consumer-grade and inexpensive research neuroheadsets allow for portability, delivering high-quality EEG signal. SmartphoneBrainScanner2 apps can be developed just like any other apps, featuring reach interface, connectivity, etc. Go, create! *How to cite SBS2* Please acknowledge the work of the SmartphoneBrainScanner2 by citing (Stopczynski et al, 2011). See also additional project related references at the homepage. A. Stopczynski, J. E. Larsen, C. Stahlhut, M. K. Petersen, & L. K. Hansen (2011), *A smartphone interface for a wireless EEG headset with real-time 3D reconstruction*, Affective Computing and Intelligent Interaction (ACII 2011), Lecture Notes in Computer Science (LNCS) 6357, Springer-Verlag Berlin Heidelberg, pp.317-318. *How to import data to Matlab* Demo scripts of how to import data to Matlab and to convert data into EEGLAB, FieldTrip, or SPM8 format are provided. Further description can be found at the Data Handling page: http://code.google.com/p/smartphonebrainscanner2/wiki/DataHandling?ts=1344340750&updated=DataHandling *Developer team* - Arkadiusz Stopczynski, DTU Informatics - Carsten Stahlhut, DTU Informatics - Michael Kai Petersen, DTU Informatics - Jakob Eg Larsen, DTU Informatics - Lars Kai Hansen, DTU Informatics *Acknowledgement* This work is supported in part by - Danish Lundbeck Foundation through Center for Integrated Molecular Brain Imaging (CIMBI ) - Danish Lundbeck Foundation through the project Real-time brain imaging by EEG - H C Ørsted Foundation - Nokia Best, Carsten -- Carsten Stahlhut Section for Cognitive Systems Department of Informatics and Mathematical Modelling Richard Petersens Plads, Building 321 Technical University of Denmark DK-2800 Kongens Lyngby, Denmark -------------- next part -------------- An HTML attachment was scrubbed... URL: From stan.vanpelt at fcdonders.ru.nl Wed Aug 8 16:14:38 2012 From: stan.vanpelt at fcdonders.ru.nl (Stan van Pelt) Date: Wed, 8 Aug 2012 16:14:38 +0200 (CEST) Subject: [FieldTrip] Fwd: source localization - MNI coordinates In-Reply-To: <1193658750.1278539.1344433666547.JavaMail.root@indus.zimbra.ru.nl> Message-ID: <1326953416.73376.1344435278179.JavaMail.root@sculptor.zimbra.ru.nl> Hi Alina, I forward your question to the FT discussion list, because I do not have experience with interpreting the source localization coordinate values. e.g., I do not know if they are actually in MNI space. ----- Doorgestuurd bericht ----- > Van: "A.S. Peter (Alina)" > Hi Stan, > > by visual inspection, I always thought that my source localization was > quite ok, but now I actually looked with MRICron where the spm > coordinates (=MNI coordinates, right?) are that fieldtrip outputs, and > some are near occipital regions but outside the brain... I am puzzled. > I tried linear ft_volumenormalize and also nonlinear, the MNI > coordinates are somewhat different but the problem remains. > > I then went back to look at SAM which we did first and it seems there > the sources were localized somewhat deeper/nicer than with LCMV > anyhow, which also tells me it's probably not a fiducial problem. I > think it may be partly my code (i.e. what I basically took from Avgis > and looks like yours) but also partly conversion to MNI causing > problems. Did you ever try to see the MNI coordinates for your data? > > Best, > > Alina -- Stan van Pelt Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen Kapittelweg 29, 6525 EN Nijmegen, Netherlands E-mail: stan.vanpelt at donders.ru.nl Website: www.ru.nl/donders/ Phone: (+31) (0)24 36 10981 Fax: (+31) (0)24 36 10989 From polomacnenad at gmail.com Thu Aug 9 11:56:13 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Thu, 9 Aug 2012 11:56:13 +0200 Subject: [FieldTrip] ordinal numbers of bad trials Message-ID: Hi, I would like to know is there an option to save ordinal number of bad trials after the artifact rejection step. I noticed that in all artifact rejection methods(ft_rejectvisual or ft_artifact_zvalue) I end up with structure "variable_name.cfg.artifact"which consist time points of the bad segments but I cannot find ordinal numbers of it. Kind regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From cornabel at googlemail.com Thu Aug 9 13:54:59 2012 From: cornabel at googlemail.com (cornelius abel) Date: Thu, 09 Aug 2012 13:54:59 +0200 Subject: [FieldTrip] ordinal numbers of bad trials In-Reply-To: References: Message-ID: <5023A513.7090208@googlemail.com> Hi Nenad, for the reject visual function i have a work around to this problem. First you have to add a column to the data.trialinfo table with a unique trial ID (in my case column 6). Then you can compare the output trialinfo of the rejectvisual function (dummy.trialinfo) to the original trialinfo and identify the rejected trials. In VisBadTrials you then have the unique trial ID of the rejected trials. Have a look at the code snip: ... dummy = ft_rejectvisual(cfg,data); VisBadTrials=data.trialinfo((find(~ismember(data.trialinfo(:,6),dummy.trialinfo(:,6)))),6)'; VisBadTrials=unique(VisBadTrials); Cornelius Am 09.08.2012 11:56, schrieb Nenad Polomac: > Hi, > > I would like to know is there an option to save ordinal number of bad > trials after the artifact rejection step. I noticed that in all > artifact rejection methods(ft_rejectvisual or ft_artifact_zvalue) I > end up with structure "variable_name.cfg.artifact"which consist time > points of the bad segments but I cannot find ordinal numbers of it. > > Kind regards! > > Nenad > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From Markus.Butz at uni-duesseldorf.de Thu Aug 9 17:24:48 2012 From: Markus.Butz at uni-duesseldorf.de (Markus Butz) Date: Thu, 09 Aug 2012 16:24:48 +0100 Subject: [FieldTrip] ft_rejectvisual: channel display Message-ID: <7570de0d8d01.5023e450@uni-duesseldorf.de> Dear list After defining 1 sec trials from continuous CTF data I want to use ft_rejectvisual using the following bit of code: cfg.method = 'trial'; chansel = ft_channelselection({'M*','EOG*','-B*','-P*','-R*','-Q*'}, data.label); cfg.channel = chansel; cfg.keepchannel = 'no'; data = ft_rejectvisual(cfg,data); What I don't understand is, why all channels are displayed in the layout, while only the selected ones show data and are saved in data.label (after going through >1 trials). However, when recalling "data = ft_rejectvisual(cfg,data);" only the selected ones are displayed in the layout with data. Does anyone has an explanation/fix for this, ie a solution that solely selected channels are displayed with the first call of ft_rejectvisual? Cheers Markus -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Fri Aug 10 14:43:23 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Fri, 10 Aug 2012 14:43:23 +0200 Subject: [FieldTrip] ordinal numbers of bad trials Message-ID: Dear Cornelius, Thank you very much for this cool solution! Kind regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From sauer.mpih at googlemail.com Fri Aug 10 15:49:47 2012 From: sauer.mpih at googlemail.com (Andreas Sauer) Date: Fri, 10 Aug 2012 15:49:47 +0200 Subject: [FieldTrip] ft_qualitycheck Message-ID: Dear all, I tried to check my MEG (CTF) datasets with ft_qualitycheck. Unfortunately it didn't work. An error occurs in the boxplot.m function in line 2020, where the data should be copied into the dataset. But for me, there is no dataset(). My question is whether this is due to my dataset-information or due to an error in the script? An how to solve this issue? Any help is highly appreciated! Thanks! Best regards, Andreas -- Andreas Sauer Max Planck Institute for Brain Research Department of Neurophysiology Deutschordenstraße 46 60528 Frankfurt am Main Germany T: +49 69 96769 278 F: +49 69 96769 327 Email: sauer.mpih at gmail.com www.brain.mpg.de -------------- next part -------------- An HTML attachment was scrubbed... URL: From a.stolk at fcdonders.ru.nl Fri Aug 10 17:17:06 2012 From: a.stolk at fcdonders.ru.nl (Stolk, A.) Date: Fri, 10 Aug 2012 17:17:06 +0200 (CEST) Subject: [FieldTrip] ft_qualitycheck In-Reply-To: Message-ID: <1868283926.90397.1344611826092.JavaMail.root@sculptor.zimbra.ru.nl> Hi Andreas, The boxplot function is a matlab statistics toolbox function. I do not see why it throws an error here. However, I did find a problem  related to the dataset history file (.hist) which contains details of the recording itself. I will fix this issue  right away. If you are still getting the same error  with tomorrow's version of FT, please give me shout. And preferably  provide some more details about the error, and how you are calling the function. best, Arjen ----- Oorspronkelijk bericht ----- > Van: "Andreas Sauer" > Aan: fieldtrip at science.ru.nl > Verzonden: Vrijdag 10 augustus 2012 15:49:47 > Onderwerp: [FieldTrip] ft_qualitycheck > Dear all, > I tried to check my MEG (CTF) datasets with ft_qualitycheck. > Unfortunately it didn't work. An error occurs in the boxplot.m > function in line 2020, where the data should be copied into the > dataset. But for me, there is no dataset(). > My question is whether this is due to my dataset-information or due to > an error in the script? An how to solve this issue? > Any help is highly appreciated! > Thanks! > Best regards, > Andreas > -- > Andreas Sauer > Max Planck Institute for Brain Research > Department of Neurophysiology > Deutschordenstraße 46 > 60528 Frankfurt am Main > Germany > T: +49 69 96769 278 > F: +49 69 96769 327 > Email: sauer.mpih at gmail.com > www.brain.mpg.de > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From sauer.mpih at googlemail.com Mon Aug 13 11:39:23 2012 From: sauer.mpih at googlemail.com (Andreas Sauer) Date: Mon, 13 Aug 2012 11:39:23 +0200 Subject: [FieldTrip] ft_qualitycheck In-Reply-To: <1868283926.90397.1344611826092.JavaMail.root@sculptor.zimbra.ru.nl> References: <1868283926.90397.1344611826092.JavaMail.root@sculptor.zimbra.ru.nl> Message-ID: Dear Arjen, thanks for your replay! I just tried with the current version of fieldtrip (20120812) but I still get this error message: ??? Undefined function or variable 'dataset'. Error in ==> boxplot>computeBoxLocation at 2020 boxValDs = dataset(); Error in ==> boxplot at 312 [boxVal,boxValPlot,boxIdx] = computeBoxLocation(xSorted,... Error in ==> ft_qualitycheck>draw_figure at 547 boxplot(h.HmotionAxes, hmotions, 'orientation', 'horizontal', 'notch', 'on'); Error in ==> ft_qualitycheck at 296 draw_figure(info, temp_timelock, temp_freq, temp_summary, temp_headpos, toi); I call the function ft_qualitycheck as described in the tutorial. cfg.dataset = subject_xy.ds; ft_qualitycheck(cfg). It runs the analysis, it writes the .mat-file but then it throws the error message. So to me, it seems that there is some information (the variable 'dataset') missing. But I don't know whether this is due to our recording or an error in one of the functions... Best, Andreas 2012/8/10 Stolk, A. > Hi Andreas, > > > > The boxplot function is a matlab statistics toolbox function. I do not see > why it throws an error here. However, I did find a problem related to the > dataset history file (.hist) which contains details of the recording > itself. I will fix this issue right away. If you are still getting the same > error with tomorrow's version of FT, please give me shout. And > preferably provide some more details about the error, and how you are > calling the function. > > > > best, > > Arjen > > > > > ------------------------------ > > *Van: *"Andreas Sauer" > *Aan: *fieldtrip at science.ru.nl > *Verzonden: *Vrijdag 10 augustus 2012 15:49:47 > *Onderwerp: *[FieldTrip] ft_qualitycheck > > > Dear all, > > I tried to check my MEG (CTF) datasets with ft_qualitycheck. Unfortunately > it didn't work. An error occurs in the boxplot.m function in line 2020, > where the data should be copied into the dataset. But for me, there is no > dataset(). > > My question is whether this is due to my dataset-information or due to an > error in the script? An how to solve this issue? > > Any help is highly appreciated! > > Thanks! > > Best regards, > > Andreas > > > -- > Andreas Sauer > Max Planck Institute for Brain Research > Department of Neurophysiology > Deutschordenstraße 46 > 60528 Frankfurt am Main > Germany > > T: +49 69 96769 278 > F: +49 69 96769 327 > Email: sauer.mpih at gmail.com > www.brain.mpg.de > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- Dipl.-Psych. Andreas Sauer Max Planck Institute for Brain Research Deutschordenstraße 46 60528 Frankfurt am Main Germany T: +49 69 96769 278 F: +49 69 96769 327 Email: sauer.mpih at gmail.com www.brain.mpg.de -------------- next part -------------- An HTML attachment was scrubbed... URL: From a.stolk at fcdonders.ru.nl Mon Aug 13 13:19:11 2012 From: a.stolk at fcdonders.ru.nl (Stolk, A.) Date: Mon, 13 Aug 2012 13:19:11 +0200 (CEST) Subject: [FieldTrip] ft_qualitycheck In-Reply-To: Message-ID: <1337301516.103006.1344856751003.JavaMail.root@sculptor.zimbra.ru.nl> Hi Andreas, Thanks for testing with the newest version. My suspicion that your bug pertains to the matlab function 'boxplot' has grown. Boxplot calls another function, 'dataset', that does exist in my version of matlab (2010b). To be more precise, it is located in the /toolbox/shared/statslib/@dataset/ directory. In my version of boxplot the line 'boxValDs = dataset();' is at line 2021 and not at line 2020. Please therefore check if your boxplot version, or your matlab version in general, is not corrupted by performing the following command: boxplot(1). Hope this helps, Arjen ----- Oorspronkelijk bericht ----- > Van: "Andreas Sauer" > Aan: "FieldTrip discussion list" > Verzonden: Maandag 13 augustus 2012 11:39:23 > Onderwerp: Re: [FieldTrip] ft_qualitycheck > Dear Arjen, > thanks for your replay! > I just tried with the current version of fieldtrip (20120812) but I > still get this error message: > ??? Undefined function or variable 'dataset'. > Error in ==> boxplot>computeBoxLocation at 2020 > boxValDs = dataset(); > Error in ==> boxplot at 312 > [boxVal,boxValPlot,boxIdx] = computeBoxLocation(xSorted,... > Error in ==> ft_qualitycheck>draw_figure at 547 >   boxplot(h.HmotionAxes, hmotions, 'orientation', 'horizontal', > 'notch', 'on'); > Error in ==> ft_qualitycheck at 296 >       draw_figure(info, temp_timelock, temp_freq, temp_summary, > temp_headpos, toi); > I call the function ft_qualitycheck as described in the tutorial. > cfg.dataset = subject_xy.ds; ft_qualitycheck(cfg). It runs the > analysis, it writes the .mat-file but then it throws the error > message. So to me, it seems that there is some information (the > variable 'dataset') missing. But I don't know whether this is due to > our recording or an error in one of the functions... > Best, > Andreas > 2012/8/10 Stolk, A. < a.stolk at fcdonders.ru.nl > > > Hi Andreas, > >   > > The boxplot function is a matlab statistics toolbox function. I do > > not > > see why it throws an error here. However, I did find a > > problem related > > to the dataset history file (.hist) which contains details of the > > recording itself. I will fix this issue right away. If you are still > > getting the same error with tomorrow's version of FT, please give me > > shout. And preferably provide some more details about the error, and > > how you are calling the function. > >   > > best, > > Arjen > >   > > > Van: "Andreas Sauer" < sauer.mpih at googlemail.com > > > > Aan: fieldtrip at science.ru.nl > > > Verzonden: Vrijdag 10 augustus 2012 15:49:47 > > > Onderwerp: [FieldTrip] ft_qualitycheck > > > Dear all, > > > I tried to check my MEG (CTF) datasets with ft_qualitycheck. > > > Unfortunately it didn't work. An error occurs in the boxplot.m > > > function in line 2020, where the data should be copied into the > > > dataset. But for me, there is no dataset(). > > > My question is whether this is due to my dataset-information or > > > due > > > to > > > an error in the script? An how to solve this issue? > > > Any help is highly appreciated! > > > Thanks! > > > Best regards, > > > Andreas > > > -- > > > Andreas Sauer > > > Max Planck Institute for Brain Research > > > Department of Neurophysiology > > > Deutschordenstraße 46 > > > 60528 Frankfurt am Main > > > Germany > > > T: +49 69 96769 278 > > > F: +49 69 96769 327 > > > Email: sauer.mpih at gmail.com > > > www.brain.mpg.de > > > _______________________________________________ > > > fieldtrip mailing list > > > fieldtrip at donders.ru.nl > > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- > Dipl.-Psych. Andreas Sauer > Max Planck Institute for Brain Research > Deutschordenstraße 46 > 60528 Frankfurt am Main > Germany > T: +49 69 96769 278 > F: +49 69 96769 327 > Email: sauer.mpih at gmail.com > www.brain.mpg.de > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From eelke.spaak at donders.ru.nl Mon Aug 13 13:30:36 2012 From: eelke.spaak at donders.ru.nl (Eelke Spaak) Date: Mon, 13 Aug 2012 13:30:36 +0200 Subject: [FieldTrip] ft_qualitycheck In-Reply-To: <1337301516.103006.1344856751003.JavaMail.root@sculptor.zimbra.ru.nl> References: <1337301516.103006.1344856751003.JavaMail.root@sculptor.zimbra.ru.nl> Message-ID: Hi Andreas, In addition to Arjen's comments, it might also be wise to execute 'which boxplot' on the Matlab command prompt, to see whether you have a version of boxplot on your path that is shadowing the stats toolbox version. Best, Eelke On 13 August 2012 13:19, Stolk, A. wrote: > Hi Andreas, > > > > Thanks for testing with the newest version. My suspicion that your bug > pertains to the matlab function 'boxplot' has grown. Boxplot calls another > function, 'dataset', that does exist in my version of matlab (2010b). To be > more precise, it is located in the /toolbox/shared/statslib/@dataset/ > directory. > > > > In my version of boxplot the line 'boxValDs = dataset();' is at line 2021 > and not at line 2020. Please therefore check if your boxplot version, or > your matlab version in general, is not corrupted by performing the following > command: boxplot(1). > > > > Hope this helps, > > Arjen > > ________________________________ > > Van: "Andreas Sauer" > Aan: "FieldTrip discussion list" > Verzonden: Maandag 13 augustus 2012 11:39:23 > Onderwerp: Re: [FieldTrip] ft_qualitycheck > > > Dear Arjen, > > thanks for your replay! > > I just tried with the current version of fieldtrip (20120812) but I still > get this error message: > > > ??? Undefined function or variable 'dataset'. > > Error in ==> boxplot>computeBoxLocation at 2020 > boxValDs = dataset(); > > Error in ==> boxplot at 312 > [boxVal,boxValPlot,boxIdx] = computeBoxLocation(xSorted,... > > Error in ==> ft_qualitycheck>draw_figure at 547 > boxplot(h.HmotionAxes, hmotions, 'orientation', 'horizontal', 'notch', > 'on'); > > Error in ==> ft_qualitycheck at 296 > draw_figure(info, temp_timelock, temp_freq, temp_summary, > temp_headpos, toi); > > > I call the function ft_qualitycheck as described in the tutorial. > cfg.dataset = subject_xy.ds; ft_qualitycheck(cfg). It runs the analysis, it > writes the .mat-file but then it throws the error message. So to me, it > seems that there is some information (the variable 'dataset') missing. But I > don't know whether this is due to our recording or an error in one of the > functions... > > Best, > > Andreas > > > > > 2012/8/10 Stolk, A. >> >> Hi Andreas, >> >> >> >> The boxplot function is a matlab statistics toolbox function. I do not see >> why it throws an error here. However, I did find a problem related to the >> dataset history file (.hist) which contains details of the recording itself. >> I will fix this issue right away. If you are still getting the same error >> with tomorrow's version of FT, please give me shout. And preferably provide >> some more details about the error, and how you are calling the function. >> >> >> >> best, >> >> Arjen >> >> >> >> >> ________________________________ >> >> Van: "Andreas Sauer" >> Aan: fieldtrip at science.ru.nl >> Verzonden: Vrijdag 10 augustus 2012 15:49:47 >> Onderwerp: [FieldTrip] ft_qualitycheck >> >> >> Dear all, >> >> I tried to check my MEG (CTF) datasets with ft_qualitycheck. Unfortunately >> it didn't work. An error occurs in the boxplot.m function in line 2020, >> where the data should be copied into the dataset. But for me, there is no >> dataset(). >> >> My question is whether this is due to my dataset-information or due to an >> error in the script? An how to solve this issue? >> >> Any help is highly appreciated! >> >> Thanks! >> >> Best regards, >> >> Andreas >> >> >> -- >> Andreas Sauer >> Max Planck Institute for Brain Research >> Department of Neurophysiology >> Deutschordenstraße 46 >> 60528 Frankfurt am Main >> Germany >> >> T: +49 69 96769 278 >> F: +49 69 96769 327 >> Email: sauer.mpih at gmail.com >> www.brain.mpg.de >> >> >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip >> >> >> >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > -- > Dipl.-Psych. Andreas Sauer > Max Planck Institute for Brain Research > Deutschordenstraße 46 > 60528 Frankfurt am Main > Germany > > T: +49 69 96769 278 > F: +49 69 96769 327 > Email: sauer.mpih at gmail.com > www.brain.mpg.de > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From maximilien.chaumon at gmail.com Mon Aug 13 14:16:56 2012 From: maximilien.chaumon at gmail.com (Maximilien Chaumon) Date: Mon, 13 Aug 2012 14:16:56 +0200 Subject: [FieldTrip] [Eeglablist] nested hypothesis testing to decide whether to use one or two dipoles to fit a component In-Reply-To: References: Message-ID: Hello eeglab & Fieldtrip, I'm trying to find out if it would be possible to use a nested hypothesis testing approach to decide whether to use a one or two dipole model while estimating components' dipole locations. The rationale I would like to follow is this: with two dipoles, we will always obtain a better fit than with one dipole, but the decrease in sum of squared errors (SSE) should follow a F distribution with k (= Nparameters_2dipoles - Nparameters_1dipole) degrees of freedom. If the decrease in SSE is greater than what would be expected under this F distribution, then we decide that 2 dipoles provide a sufficiently better fit and decide using them. I asked this question to eeglablist and Scott pointed out that it is difficult/impossible(?) to determine if the second dipole fits actual interesting data or just noise introduced by the imperfect head model. Christian then said it'd be worth a shot, and I agree, so here I am again with two questions, or two confirmations: 1) *How many parameters are estimated in ft_dipolefitting.m ?* specially in the case of 2 dipoles. If I count correctly, we estimate 6 parameters for one dipole, and, depending on whether the orientation has to be the same in the 2 dipoles, one (amplitude) or three (amplitude and orientation) more. 2) *Can I assimilate the relative residual variance to a SSE?* the function rv.m does this: rv = sum((d1-d2).^2) ./ sum(d1.^2); So that seems to be a sum of squared errors divided by the variance of the original data. So if I multiply the rv by the sum squared component map, I should get it, right? Thanks a lot! Max 2012/8/11 Christian Kothe > I can only speak from my armchair here, but it sounds like it should be > worth a try - even if you don't get the # of parameters exactly right it > will probably give you at least some level of complexity control in > whatever the range of validity is. If it works, it may inspire follow-up > work (e.g., Bayesian model selection or likelihood ratio tests). > > The number of parameters for a 2-dipole model seems to be 3 (xyz) + 4 (2x > the orientation parameters). Not sure about the momentum, though - you > might look up the place where the actual function minimization is being > performed in dipfit (fminunc call?) and see whether these are being > optimized together with the others. > > Christian > > On Aug 10, 2012, at 3:29 PM, Scott Makeig wrote: > > MAX - Unfortunately, in general using two dipoles rather than one will > ~always improve the fit. Even if the source is a pure single dipole, a > second dipole can be used to correct for noise or errors in the forward > head model. This is less always the case for the constrained > spatially-symmetric dipole pairs allowed by dipfit(). However, we have not > thought of an optimal way to decide between using one or (occasionally) two > dipoles to fit e.g. maps of ICA brain sources. The goal would be to decide > whether the two-dipole version is fitting noise/forward model error vs > actual bilateral source generation... > > Scott Makeig > > On Thu, Aug 9, 2012 at 1:54 AM, Maximilien Chaumon < > maximilien.chaumon at gmail.com> wrote: > >> Hello all, >> >> When fitting dipoles to components, we are all sooner or later puzzled by >> the question whether to use one or two symmetrical dipoles. >> >> Would it be correct to put the problems in terms of a nested hypothesis >> testing? >> >> We are fitting a scalp map with one or two parameters and get a residual >> variance after the fit. >> Could we not use this residual variance as a measure of the SSE and >> compute a F statistic to decide whether to use the more complex (with two >> dipoles) or simpler (with one dipole) of two nested models? >> If yes, then how would we decide on the number of degrees of freedom? How >> many free parameters do we have in each case? x,y,z,and two orientations >> per dipole? how does the imposed symmetry affect that number? Could we >> really map residual variance to SSE? How many "observations" do we have in >> that case (see formula below)? >> >> I found this formula, for F: >> F = (SSEF-SSER)/ (kF-kR) / ((1-SSEF)/(N-kF-1)) >> where >> SSE is sum of squared errors, >> k is numbers of parameters, >> N is number of observations (? what in our case?) >> F and R indices for full and reduced model respectively (in our case two >> and one dipole). >> >> >> Thanks a lot for any comment! >> Best, >> Max >> >> dipfit >> >> >> >> >> >> _______________________________________________ >> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html >> To unsubscribe, send an empty email to >> eeglablist-unsubscribe at sccn.ucsd.edu >> For digest mode, send an email with the subject "set digest mime" to >> eeglablist-request at sccn.ucsd.edu >> > > > > -- > Scott Makeig, Research Scientist and Director, Swartz Center for > Computational Neuroscience, Institute for Neural Computation; Prof. of > Neurosciences (Adj.), University of California San Diego, La Jolla CA > 92093-0559, http://sccn.ucsd.edu/~scott > > _______________________________________________ > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html > To unsubscribe, send an empty email to > eeglablist-unsubscribe at sccn.ucsd.edu > For digest mode, send an email with the subject "set digest mime" to > eeglablist-request at sccn.ucsd.edu > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From sauer.mpih at googlemail.com Mon Aug 13 14:59:09 2012 From: sauer.mpih at googlemail.com (Andreas Sauer) Date: Mon, 13 Aug 2012 14:59:09 +0200 Subject: [FieldTrip] ft_qualitycheck In-Reply-To: References: <1337301516.103006.1344856751003.JavaMail.root@sculptor.zimbra.ru.nl> Message-ID: Dear Eelke, dear Arjen, thanks for helping! Here, I analyze data on a cluster with Matlab 2008b. Although I found a folder "@dataset" in /toolbox/shared/statslib/, unfortunately there is no such function in it. Since I found the function in my local Matlab, which is version 2010a, I guess that this is a feature of newer Matlab versions. I downloaded one dataset to my local machine and with Matlab 2010a it worked... I am not sure whether it might be good to note that in the tutorial... Best, Andreas 2012/8/13 Eelke Spaak > Hi Andreas, > > In addition to Arjen's comments, it might also be wise to execute > 'which boxplot' on the Matlab command prompt, to see whether you have > a version of boxplot on your path that is shadowing the stats toolbox > version. > > Best, > Eelke > > On 13 August 2012 13:19, Stolk, A. wrote: > > Hi Andreas, > > > > > > > > Thanks for testing with the newest version. My suspicion that your bug > > pertains to the matlab function 'boxplot' has grown. Boxplot calls > another > > function, 'dataset', that does exist in my version of matlab (2010b). To > be > > more precise, it is located in the /toolbox/shared/statslib/@dataset/ > > directory. > > > > > > > > In my version of boxplot the line 'boxValDs = dataset();' is at line 2021 > > and not at line 2020. Please therefore check if your boxplot version, or > > your matlab version in general, is not corrupted by performing the > following > > command: boxplot(1). > > > > > > > > Hope this helps, > > > > Arjen > > > > ________________________________ > > > > Van: "Andreas Sauer" > > Aan: "FieldTrip discussion list" > > Verzonden: Maandag 13 augustus 2012 11:39:23 > > Onderwerp: Re: [FieldTrip] ft_qualitycheck > > > > > > Dear Arjen, > > > > thanks for your replay! > > > > I just tried with the current version of fieldtrip (20120812) but I still > > get this error message: > > > > > > ??? Undefined function or variable 'dataset'. > > > > Error in ==> boxplot>computeBoxLocation at 2020 > > boxValDs = dataset(); > > > > Error in ==> boxplot at 312 > > [boxVal,boxValPlot,boxIdx] = computeBoxLocation(xSorted,... > > > > Error in ==> ft_qualitycheck>draw_figure at 547 > > boxplot(h.HmotionAxes, hmotions, 'orientation', 'horizontal', 'notch', > > 'on'); > > > > Error in ==> ft_qualitycheck at 296 > > draw_figure(info, temp_timelock, temp_freq, temp_summary, > > temp_headpos, toi); > > > > > > I call the function ft_qualitycheck as described in the tutorial. > > cfg.dataset = subject_xy.ds; ft_qualitycheck(cfg). It runs the analysis, > it > > writes the .mat-file but then it throws the error message. So to me, it > > seems that there is some information (the variable 'dataset') missing. > But I > > don't know whether this is due to our recording or an error in one of the > > functions... > > > > Best, > > > > Andreas > > > > > > > > > > 2012/8/10 Stolk, A. > >> > >> Hi Andreas, > >> > >> > >> > >> The boxplot function is a matlab statistics toolbox function. I do not > see > >> why it throws an error here. However, I did find a problem related to > the > >> dataset history file (.hist) which contains details of the recording > itself. > >> I will fix this issue right away. If you are still getting the same > error > >> with tomorrow's version of FT, please give me shout. And preferably > provide > >> some more details about the error, and how you are calling the function. > >> > >> > >> > >> best, > >> > >> Arjen > >> > >> > >> > >> > >> ________________________________ > >> > >> Van: "Andreas Sauer" > >> Aan: fieldtrip at science.ru.nl > >> Verzonden: Vrijdag 10 augustus 2012 15:49:47 > >> Onderwerp: [FieldTrip] ft_qualitycheck > >> > >> > >> Dear all, > >> > >> I tried to check my MEG (CTF) datasets with ft_qualitycheck. > Unfortunately > >> it didn't work. An error occurs in the boxplot.m function in line 2020, > >> where the data should be copied into the dataset. But for me, there is > no > >> dataset(). > >> > >> My question is whether this is due to my dataset-information or due to > an > >> error in the script? An how to solve this issue? > >> > >> Any help is highly appreciated! > >> > >> Thanks! > >> > >> Best regards, > >> > >> Andreas > >> > >> > >> -- > >> Andreas Sauer > >> Max Planck Institute for Brain Research > >> Department of Neurophysiology > >> Deutschordenstraße 46 > >> 60528 Frankfurt am Main > >> Germany > >> > >> T: +49 69 96769 278 > >> F: +49 69 96769 327 > >> Email: sauer.mpih at gmail.com > >> www.brain.mpg.de > >> > >> > >> _______________________________________________ > >> fieldtrip mailing list > >> fieldtrip at donders.ru.nl > >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > >> > >> > >> > >> _______________________________________________ > >> fieldtrip mailing list > >> fieldtrip at donders.ru.nl > >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > > > > > > -- > > Dipl.-Psych. Andreas Sauer > > Max Planck Institute for Brain Research > > Deutschordenstraße 46 > > 60528 Frankfurt am Main > > Germany > > > > T: +49 69 96769 278 > > F: +49 69 96769 327 > > Email: sauer.mpih at gmail.com > > www.brain.mpg.de > > > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- Dipl.-Psych. Andreas Sauer Max Planck Institute for Brain Research Deutschordenstraße 46 60528 Frankfurt am Main Germany T: +49 69 96769 278 F: +49 69 96769 327 Email: sauer.mpih at gmail.com www.brain.mpg.de -------------- next part -------------- An HTML attachment was scrubbed... URL: From akiko.ikkai at gmail.com Tue Aug 14 22:19:13 2012 From: akiko.ikkai at gmail.com (Akiko Ikkai) Date: Tue, 14 Aug 2012 16:19:13 -0400 Subject: [FieldTrip] actvsblT and average over time and/or frequency Message-ID: Hi all, I'm testing for the power difference between baseline (fixation) and memory delay periods, following the examples in "Permutation Tests > Within trial experiments" tutorial. When I set cfg.statistic = 'actvsblT'; and cfg.avgovertime = 'yes'; cfg.avgoverfreq = 'yes'; then run ft_freqstatistics, I get an error in statfun_actvsblT.m, "Inappropriate dimord for the statistics function STATFUN_ACTVSBLT.". This is because with avgovertime and/or avgoverfreq set to 'yes', cfg.dimord for data input to statfun_actvsblT is simply "chan," whereas statfun_actvsblT requires "chan_freq_time." When I comment out cfg.avgovertime AND freq, permutation runs fine. This is particular to statfun_actvsblT; other functions, such as statfun_depsamplesT, do not have this dimord structure requirements. I'm puzzled why averaging over time and/or frequency could not be a valid option for actvsblT... could someone help me understand? Thanks! Akiko -- Akiko Ikkai, Ph.D. Postdoctoral Fellow Department of Psychological and Brain Sciences Johns Hopkins University Ames Hall, 3400 N. Charles St. Baltimore, MD 21218 -------------- next part -------------- An HTML attachment was scrubbed... URL: From aler1 at gmx.de Wed Aug 15 15:47:20 2012 From: aler1 at gmx.de (alla Brodski) Date: Wed, 15 Aug 2012 15:47:20 +0200 Subject: [FieldTrip] Source analysis power unit Message-ID: <20120815134720.43010@gmx.net> Dear fieldtrip users, I am using fieldtrip for beamforming source analysis. I try to interpret the units of the source power results, which are for my data in the range of 10^-07 to 10^-10. I am using the fieldtrip version of 01.05.2012; with an older version (e.g. of 2009) I get the same numbers but in a range of 10^27 to 10^30. I have already noticed that there was a change for the singleshell method, which I am using; a new scaling factor was introduced in the beginning of 2011: http://mailman.science.ru.nl/pipermail/fieldtrip/2011-January/003367.html However, I am still wondering whether I can interpret the results in physical units or not. Can they still be interpreted in tesla /fT (as if they had the old scale)? Or am I possibly doing something wrong as 10^-07 to 10^-10 does not seem a meaningful range. Thank you very much! Alla Brodski From aler1 at gmx.de Wed Aug 15 16:07:16 2012 From: aler1 at gmx.de (alla Brodski) Date: Wed, 15 Aug 2012 16:07:16 +0200 Subject: [FieldTrip] Source analysis power unit In-Reply-To: <20120815134720.43010@gmx.net> References: <20120815134720.43010@gmx.net> Message-ID: <20120815140716.43010@gmx.net> Sorry, of course it is 10^-27 to 10^-30 with an old version -------- Original-Nachricht -------- > Datum: Wed, 15 Aug 2012 15:47:20 +0200 > Von: "alla Brodski" > An: fieldtrip at science.ru.nl > Betreff: [FieldTrip] Source analysis power unit > Dear fieldtrip users, > > I am using fieldtrip for beamforming source analysis. > I try to interpret the units of the source power results, which are for my > data in the range of 10^-07 to 10^-10. I am using the fieldtrip version of > 01.05.2012; with an older version (e.g. of 2009) I get the same numbers > but in a range of 10^27 to 10^30. > I have already noticed that there was a change for the singleshell method, > which I am using; a new scaling factor was introduced in the beginning of > 2011: > http://mailman.science.ru.nl/pipermail/fieldtrip/2011-January/003367.html > > However, I am still wondering whether I can interpret the results in > physical units or not. Can they still be interpreted in tesla /fT (as if they > had the old scale)? Or am I possibly doing something wrong as 10^-07 to > 10^-10 does not seem a meaningful range. > > Thank you very much! > > Alla Brodski > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From Paul_C at gmx.de Fri Aug 17 01:12:04 2012 From: Paul_C at gmx.de (Paul Kertscher) Date: Fri, 17 Aug 2012 01:12:04 +0200 Subject: [FieldTrip] Question about ft_dipolesimulation Message-ID: <20120816231204.162220@gmx.net> Dear fieldtrippers, I recetly asked myself what units the signal in ft_dipolsimulation had. First I thought, that it had to be a current, thus Ampére. After rechecking it in the Kybic et al. paper it obviously is an current density. If the signal simulated has an amplitude of 1, does this mean, that the current density would be 1A/m², or are these units in the function rather of an arbitrary kind? Best regards, Paul Kertscher --- Paul Kertscher (Dipl.-Phys.) Björnsonstr. 25 12163 Berlin +49/(0)30/221609359 +49/(0)1578/7839933 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Andrew.Dimitrijevic at cchmc.org Fri Aug 17 04:41:40 2012 From: Andrew.Dimitrijevic at cchmc.org (Dimitrijevic, Andrew) Date: Fri, 17 Aug 2012 02:41:40 +0000 Subject: [FieldTrip] postdoc position available ... auditory ERPs in cochlear implant users Message-ID: <1787E0F7D5DAD54E803CBB4E877A33BE026B7C90@MCEXMB3.chmccorp.cchmc.org> Dear FieldTrippers ... postdoc position available, feel free to pass along (sorry for cross posting) Postdoctoral position in auditory EPs with cochlear implant users We are seeking a highly motivated postdoctoral fellow with expertise in EEGLAB/FieldTrip to the join the Hearing Lab at the Communication Sciences Research Center (CSRC) at Cincinnati Children’s Hospital Medical Center (CCHMC). This particular project will involve psychoacoustics and high-density EEG recordings in adults and children with cochlear implants. Much of the work will involve post processing in Matlab therefore solid programming ability is essential. The ideal candidate will have a PhD in cognitive science/neuroscience/computer science/psychology/engineering or related discipline, and an interest in cochlear implants and EEG. The CSRC has a strong interdisciplinary community that has both clinicians and basic scientists. Our facilities include an EEG lab, a Neuroimaging center with research dedicated 3T MRI scanner, 275-channel MEG lab. The position is for 2 years and second year is contingent on satisfactory progress. The anticipated start date is Jan 2012. Applicants should submit a cover letter and CV, including the names and contact information of three references. Please send application materials by e-mail to: Andrew Dimitrijevic PhD Assistant Professor Communication Sciences Research Center https://csrc.cchmc.org/ andrew.dimitrijevic at cchmc.org https://csrc.cchmc.org/andrew-dimitrijevic CCHMC has a longstanding commitment to achieving diversity among faculty, staff, and students. CCHMC is an EO/AA Employer. -------------- next part -------------- An HTML attachment was scrubbed... URL: From julian.keil at gmail.com Fri Aug 17 14:50:25 2012 From: julian.keil at gmail.com (Julian Keil) Date: Fri, 17 Aug 2012 14:50:25 +0200 Subject: [FieldTrip] FIeldTrip Tutorian Videos Message-ID: <82B32158-73E9-4E12-B91A-C398B996EB34@gmail.com> Dear FieldTrippers, a while ago I held a little workshop at the BRAMS Institute in Montreal covering the basic steps in FieldTrip. I recorded the whole session and finally got around to put the videos online. Please feel free to check it out at or point any new users to it: http://vimeo.com/user11934546/videos The accompanying Matlab script can be found here: http://pastebin.com/yskmVKAh Please also tell me if I made any errors or you feel that I have explained something incorrectly. Best, Julian ******************************************** Dr. Julian Keil International Laboratory for Brain Music and Sound Research (BRAMS) Pavillon 1430 Mont-Royal Université de Montréal Montréal, Québec Canada, H2V 4P3 julian.keil at gmail.com +1-514-343-6111-29653 www.brams.org -------------- next part -------------- An HTML attachment was scrubbed... URL: From pgoodin at swin.edu.au Sun Aug 19 13:17:16 2012 From: pgoodin at swin.edu.au (Peter Goodin) Date: Sun, 19 Aug 2012 11:17:16 +0000 Subject: [FieldTrip] Wavelet advice Message-ID: Hi Fieldtrip list, I'm attempting to use wavelets to analyse some data collected with a neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is 1.6 seconds) but am new to time frequency analysis. I'm interested in lower frequency bins (~4 to 20 Hz). The config settings I've been playing with are as follows: cfg.method = 'wavelet'; cfg.channel = 'MEG'; cfg.keeptrials = 'yes'; cfg.foi = [4:1:20] cfg.toi = [-.1:.01:1.5] cfg.width = x Using a default width of 7, I'm getting large boundary effects leading to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. I've played with the width and have found that a width of 1 gives only a small amount of boundary effect at the extreme edges and shows a large increase in the frequencies I'm interested in at a time point which corresponds quite nicely with the ERF data also analysed. Having read the Tallon-Baudry (1999) article, material from the tutorials and items from the mailing list, I understand that by using a wavelet width of 1 I've biased my results towards higher temporal vs. spectral resolution, but considering the low range of frequencies I'm interest in, is this a problem? Does using a low width with my cfg.foi set as it is just lead to a large amount of redundant data? Advice on this subject would be greatly appreciated. Peter. -------------- next part -------------- An HTML attachment was scrubbed... URL: From smoratti at psi.ucm.es Sun Aug 19 14:19:26 2012 From: smoratti at psi.ucm.es (smoratti at psi.ucm.es) Date: Sun, 19 Aug 2012 14:19:26 +0200 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: Message-ID: Dear Peter, I think you should use 5 cycles as minimum to get stable results. One cycle is not much to estimate the power of your time-freq of interest. Best, Stephan ________________________________________________________ Stephan Moratti, PhD see also: http://web.me.com/smoratti/ Universidad Complutense de Madrid Facultad de Psicología Departamento de Psicología Básica I Campus de Somosaguas 28223 Pozuelo de Alarcón (Madrid) Spain and Center for Biomedical Technology Laboratory for Cognitive and Computational Neuroscience Parque Científico y Tecnológico de la Universidad Politecnica de Madrid Campus Montegancedo 28223 Pozuelo de Alarcón (Madrid) Spain email: smoratti at psi.ucm.es Tel.: +34 679219982 El 19/08/2012, a las 13:17, Peter Goodin escribió: > Hi Fieldtrip list, > > I'm attempting to use wavelets to analyse some data collected with a neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is 1.6 seconds) but am new to time frequency analysis. I'm interested in lower frequency bins (~4 to 20 Hz). The config settings I've been playing with are as follows: > > cfg.method = 'wavelet'; > cfg.channel = 'MEG'; > cfg.keeptrials = 'yes'; > cfg.foi = [4:1:20] > cfg.toi = [-.1:.01:1.5] > cfg.width = x > > Using a default width of 7, I'm getting large boundary effects leading to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. I've played with the width and have found that a width of 1 gives only a small amount of boundary effect at the extreme edges and shows a large increase in the frequencies I'm interested in at a time point which corresponds quite nicely with the ERF data also analysed. > > Having read the Tallon-Baudry (1999) article, material from the tutorials and items from the mailing list, I understand that by using a wavelet width of 1 I've biased my results towards higher temporal vs. spectral resolution, but considering the low range of frequencies I'm interest in, is this a problem? Does using a low width with my cfg.foi set as it is just lead to a large amount of redundant data? > > Advice on this subject would be greatly appreciated. > > Peter. > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From sherrykhan78 at gmail.com Sun Aug 19 22:58:21 2012 From: sherrykhan78 at gmail.com (Sheraz Khan) Date: Sun, 19 Aug 2012 16:58:21 -0400 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: Message-ID: Hi, You can always try variables number of cycles may be start with 3 and then go to 5. Sheraz Martinos Center MGH/MIT/Harvard On Aug 19, 2012 7:17 AM, "Peter Goodin" wrote: > Hi Fieldtrip list, > > I'm attempting to use wavelets to analyse some data collected with a > neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is > 1.6 seconds) but am new to time frequency analysis. I'm interested in lower > frequency bins (~4 to 20 Hz). The config settings I've been playing with > are as follows: > > cfg.method = 'wavelet'; > cfg.channel = 'MEG'; > cfg.keeptrials = 'yes'; > cfg.foi = [4:1:20] > cfg.toi = [-.1:.01:1.5] > cfg.width = x > > Using a default width of 7, I'm getting large boundary effects leading > to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. > I've played with the width and have found that a width of 1 gives only a > small amount of boundary effect at the extreme edges and shows a large > increase in the frequencies I'm interested in at a time point which > corresponds quite nicely with the ERF data also analysed. > > Having read the Tallon-Baudry (1999) article, material from the > tutorials and items from the mailing list, I understand that by using a > wavelet width of 1 I've biased my results towards higher temporal vs. > spectral resolution, but considering the low range of frequencies I'm > interest in, is this a problem? Does using a low width with my cfg.foi set > as it is just lead to a large amount of redundant data? > > Advice on this subject would be greatly appreciated. > > Peter. > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pgoodin at swin.edu.au Mon Aug 20 01:20:49 2012 From: pgoodin at swin.edu.au (Peter Goodin) Date: Sun, 19 Aug 2012 23:20:49 +0000 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: , Message-ID: Thanks for all the replies so far. Re-reading my previous question I appear to have been a bit vague to the overall problem. The problem is I've used wavelets widths starting from 1 and increased to the default 7, but only a value of 1 gives me any usable data around 4Hz. What confuses me is that Kaplan et al, 2011 (http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001267) were able to get theta band activity using epochs of 1s (200ms pre 800ms post) without apparent boundary effects at a width of 5, but when I tried the same thing (including zero padding), the results were not usable. Again, any advice and explanations as to why 1. a value of 1 is not usable and 2. why I may be getting these boundary effects would be greatly appreciated. Peter ________________________________ From: fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl] on behalf of Sheraz Khan [sherrykhan78 at gmail.com] Sent: Monday, 20 August 2012 6:58 AM To: FieldTrip discussion list Subject: Re: [FieldTrip] Wavelet advice Hi, You can always try variables number of cycles may be start with 3 and then go to 5. Sheraz Martinos Center MGH/MIT/Harvard On Aug 19, 2012 7:17 AM, "Peter Goodin" > wrote: Hi Fieldtrip list, I'm attempting to use wavelets to analyse some data collected with a neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is 1.6 seconds) but am new to time frequency analysis. I'm interested in lower frequency bins (~4 to 20 Hz). The config settings I've been playing with are as follows: cfg.method = 'wavelet'; cfg.channel = 'MEG'; cfg.keeptrials = 'yes'; cfg.foi = [4:1:20] cfg.toi = [-.1:.01:1.5] cfg.width = x Using a default width of 7, I'm getting large boundary effects leading to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. I've played with the width and have found that a width of 1 gives only a small amount of boundary effect at the extreme edges and shows a large increase in the frequencies I'm interested in at a time point which corresponds quite nicely with the ERF data also analysed. Having read the Tallon-Baudry (1999) article, material from the tutorials and items from the mailing list, I understand that by using a wavelet width of 1 I've biased my results towards higher temporal vs. spectral resolution, but considering the low range of frequencies I'm interest in, is this a problem? Does using a low width with my cfg.foi set as it is just lead to a large amount of redundant data? Advice on this subject would be greatly appreciated. Peter. _______________________________________________ fieldtrip mailing list fieldtrip at donders.ru.nl http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From sherrykhan78 at gmail.com Mon Aug 20 03:29:42 2012 From: sherrykhan78 at gmail.com (Sheraz Khan) Date: Sun, 19 Aug 2012 21:29:42 -0400 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: Message-ID: You can not avoid the border effects they comes from the convolution. Only possible solution which I sometimes use when the epochs are shorter due to the design, is to do the wavelet decomposition on the continuous data, and then epoch it, this will avoid the borders from the epoch. Sheraz Martinos Center MGH/MIT/Harvard On Sun, Aug 19, 2012 at 7:20 PM, Peter Goodin wrote: > Thanks for all the replies so far. > > Re-reading my previous question I appear to have been a bit vague to the > overall problem. > > The problem is I've used wavelets widths starting from 1 and increased > to the default 7, but only a value of 1 gives me any usable data around > 4Hz. > > What confuses me is that Kaplan et al, 2011 ( > http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001267) > were able to get theta band activity using epochs of 1s (200ms pre 800ms > post) without apparent boundary effects at a width of 5, but when I tried > the same thing (including zero padding), the results were not usable. > > Again, any advice and explanations as to why 1. a value of 1 is not > usable and 2. why I may be getting these boundary effects would be greatly > appreciated. > > Peter > ------------------------------ > *From:* fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl] > on behalf of Sheraz Khan [sherrykhan78 at gmail.com] > *Sent:* Monday, 20 August 2012 6:58 AM > *To:* FieldTrip discussion list > *Subject:* Re: [FieldTrip] Wavelet advice > > Hi, > You can always try variables number of cycles may be start with 3 and then > go to 5. > Sheraz > Martinos Center > MGH/MIT/Harvard > On Aug 19, 2012 7:17 AM, "Peter Goodin" wrote: > >> Hi Fieldtrip list, >> >> I'm attempting to use wavelets to analyse some data collected with a >> neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is >> 1.6 seconds) but am new to time frequency analysis. I'm interested in lower >> frequency bins (~4 to 20 Hz). The config settings I've been playing with >> are as follows: >> >> cfg.method = 'wavelet'; >> cfg.channel = 'MEG'; >> cfg.keeptrials = 'yes'; >> cfg.foi = [4:1:20] >> cfg.toi = [-.1:.01:1.5] >> cfg.width = x >> >> Using a default width of 7, I'm getting large boundary effects leading >> to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. >> I've played with the width and have found that a width of 1 gives only a >> small amount of boundary effect at the extreme edges and shows a large >> increase in the frequencies I'm interested in at a time point which >> corresponds quite nicely with the ERF data also analysed. >> >> Having read the Tallon-Baudry (1999) article, material from the >> tutorials and items from the mailing list, I understand that by using a >> wavelet width of 1 I've biased my results towards higher temporal vs. >> spectral resolution, but considering the low range of frequencies I'm >> interest in, is this a problem? Does using a low width with my cfg.foi set >> as it is just lead to a large amount of redundant data? >> >> Advice on this subject would be greatly appreciated. >> >> Peter. >> >> >> >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip >> > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pgoodin at swin.edu.au Mon Aug 20 05:18:09 2012 From: pgoodin at swin.edu.au (Peter Goodin) Date: Mon, 20 Aug 2012 03:18:09 +0000 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: , Message-ID: Thanks for that! Much appreciated. ________________________________ From: fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl] on behalf of Sheraz Khan [sherrykhan78 at gmail.com] Sent: Monday, 20 August 2012 11:29 AM To: FieldTrip discussion list Subject: Re: [FieldTrip] Wavelet advice You can not avoid the border effects they comes from the convolution. Only possible solution which I sometimes use when the epochs are shorter due to the design, is to do the wavelet decomposition on the continuous data, and then epoch it, this will avoid the borders from the epoch. Sheraz Martinos Center MGH/MIT/Harvard On Sun, Aug 19, 2012 at 7:20 PM, Peter Goodin > wrote: Thanks for all the replies so far. Re-reading my previous question I appear to have been a bit vague to the overall problem. The problem is I've used wavelets widths starting from 1 and increased to the default 7, but only a value of 1 gives me any usable data around 4Hz. What confuses me is that Kaplan et al, 2011 (http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001267) were able to get theta band activity using epochs of 1s (200ms pre 800ms post) without apparent boundary effects at a width of 5, but when I tried the same thing (including zero padding), the results were not usable. Again, any advice and explanations as to why 1. a value of 1 is not usable and 2. why I may be getting these boundary effects would be greatly appreciated. Peter ________________________________ From: fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl] on behalf of Sheraz Khan [sherrykhan78 at gmail.com] Sent: Monday, 20 August 2012 6:58 AM To: FieldTrip discussion list Subject: Re: [FieldTrip] Wavelet advice Hi, You can always try variables number of cycles may be start with 3 and then go to 5. Sheraz Martinos Center MGH/MIT/Harvard On Aug 19, 2012 7:17 AM, "Peter Goodin" > wrote: Hi Fieldtrip list, I'm attempting to use wavelets to analyse some data collected with a neuromag system (sample rate of 500Hz, pre-trigger period is 200ms, post is 1.6 seconds) but am new to time frequency analysis. I'm interested in lower frequency bins (~4 to 20 Hz). The config settings I've been playing with are as follows: cfg.method = 'wavelet'; cfg.channel = 'MEG'; cfg.keeptrials = 'yes'; cfg.foi = [4:1:20] cfg.toi = [-.1:.01:1.5] cfg.width = x Using a default width of 7, I'm getting large boundary effects leading to a period of ~300ms (400 - 700ms post trigger) of calculated data at 4Hz. I've played with the width and have found that a width of 1 gives only a small amount of boundary effect at the extreme edges and shows a large increase in the frequencies I'm interested in at a time point which corresponds quite nicely with the ERF data also analysed. Having read the Tallon-Baudry (1999) article, material from the tutorials and items from the mailing list, I understand that by using a wavelet width of 1 I've biased my results towards higher temporal vs. spectral resolution, but considering the low range of frequencies I'm interest in, is this a problem? Does using a low width with my cfg.foi set as it is just lead to a large amount of redundant data? Advice on this subject would be greatly appreciated. Peter. _______________________________________________ fieldtrip mailing list fieldtrip at donders.ru.nl http://mailman.science.ru.nl/mailman/listinfo/fieldtrip _______________________________________________ fieldtrip mailing list fieldtrip at donders.ru.nl http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From nuria.donamayor at neuro.uni-luebeck.de Tue Aug 21 12:50:56 2012 From: nuria.donamayor at neuro.uni-luebeck.de (=?iso-8859-1?Q?Nuria_Do=F1amayor_Alonso?=) Date: Tue, 21 Aug 2012 12:50:56 +0200 Subject: [FieldTrip] source reconstruction on emfs Message-ID: <810A8E06C75EB447A8CEB73DBFD7BB0E4C03CD65AD@solaris.neuro.uni-luebeck.de> Dear fieldtrippers, I am currently analyzing a study, measured with a Neuromag 306 system, in which we stimulated the median nerve. I had never done source reconstruction with fieldtrip, so I followed the tutorial using MNE for EMF source analysis (which worked fine), only to find out that I could neither average individual sources nor perform stats, which is exactly what I need to do... So the question is: What would you guys recommend? Should I just try using the LCMV beamfomer? However, that might not be the best solution according to this thread http://mailman.science.ru.nl/pipermail/fieldtrip/2011-September/004321.html... Is there any way I could efficiently export my preprocessed/averaged data to some other software (MNE/SPM/Brainstorm...) and compute my source reconstruction and stats there? What do you guys usually do? Thanks a lot for your help, Nuria From jm.horschig at donders.ru.nl Tue Aug 21 13:24:21 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Tue, 21 Aug 2012 13:24:21 +0200 Subject: [FieldTrip] Wavelet advice In-Reply-To: References: , Message-ID: <50336FE5.30602@donders.ru.nl> Hey Peter, when doing what Sheraz suggested you should, however, keep in mind that the edges are computed using data outside your epoch, i.e. possible muscle artifacts and other stuff (e.g. stimulation) will corrupt the frequency estimation if not properly accounted for. Make sure to note that when interpreting your data. Best, Jörn On 8/20/2012 5:18 AM, Peter Goodin wrote: > Thanks for that! Much appreciated. > ------------------------------------------------------------------------ > *From:* fieldtrip-bounces at science.ru.nl > [fieldtrip-bounces at science.ru.nl] on behalf of Sheraz Khan > [sherrykhan78 at gmail.com] > *Sent:* Monday, 20 August 2012 11:29 AM > *To:* FieldTrip discussion list > *Subject:* Re: [FieldTrip] Wavelet advice > > You can not avoid the border effects they comes from the convolution. > > Only possible solution which I sometimes use when the epochs are > shorter due to the design, is to do the wavelet decomposition on > the continuous data, and then epoch it, this will avoid the borders > from the epoch. > > Sheraz > Martinos Center > MGH/MIT/Harvard > > On Sun, Aug 19, 2012 at 7:20 PM, Peter Goodin > wrote: > > Thanks for all the replies so far. > > Re-reading my previous question I appear to have been a bit vague > to the overall problem. > > The problem is I've used wavelets widths starting from 1 and > increased to the default 7, but only a value of 1 gives me any > usable data around 4Hz. > > What confuses me is that Kaplan et al, 2011 > (http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001267) > were able to get theta band activity using epochs of 1s (200ms pre > 800ms post) without apparent boundary effects at a width of 5, but > when I tried the same thing (including zero padding), the results > were not usable. > > Again, any advice and explanations as to why 1. a value of 1 is > not usable and 2. why I may be getting these boundary effects > would be greatly appreciated. > > Peter > ------------------------------------------------------------------------ > *From:* fieldtrip-bounces at science.ru.nl > > [fieldtrip-bounces at science.ru.nl > ] on behalf of Sheraz Khan > [sherrykhan78 at gmail.com ] > *Sent:* Monday, 20 August 2012 6:58 AM > *To:* FieldTrip discussion list > *Subject:* Re: [FieldTrip] Wavelet advice > > Hi, > You can always try variables number of cycles may be start with 3 > and then go to 5. > Sheraz > Martinos Center > MGH/MIT/Harvard > > On Aug 19, 2012 7:17 AM, "Peter Goodin" > wrote: > > Hi Fieldtrip list, > > I'm attempting to use wavelets to analyse some data collected > with a neuromag system (sample rate of 500Hz, pre-trigger > period is 200ms, post is 1.6 seconds) but am new to time > frequency analysis. I'm interested in lower frequency bins (~4 > to 20 Hz). The config settings I've been playing with are as > follows: > > cfg.method = 'wavelet'; > cfg.channel = 'MEG'; > cfg.keeptrials = 'yes'; > cfg.foi = [4:1:20] > cfg.toi = [-.1:.01:1.5] > cfg.width = x > > Using a default width of 7, I'm getting large boundary effects > leading to a period of ~300ms (400 - 700ms post trigger) of > calculated data at 4Hz. I've played with the width and have > found that a width of 1 gives only a small amount of boundary > effect at the extreme edges and shows a large increase in the > frequencies I'm interested in at a time point which > corresponds quite nicely with the ERF data also analysed. > > Having read the Tallon-Baudry (1999) article, material from > the tutorials and items from the mailing list, I understand > that by using a wavelet width of 1 I've biased my results > towards higher temporal vs. spectral resolution, but > considering the low range of frequencies I'm interest in, is > this a problem? Does using a low width with my cfg.foi set as > it is just lead to a large amount of redundant data? > > Advice on this subject would be greatly appreciated. > > Peter. > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group FieldTrip Development Team P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From r.oostenveld at donders.ru.nl Tue Aug 21 21:56:20 2012 From: r.oostenveld at donders.ru.nl (Robert Oostenveld) Date: Tue, 21 Aug 2012 21:56:20 +0200 Subject: [FieldTrip] source reconstruction on emfs In-Reply-To: <810A8E06C75EB447A8CEB73DBFD7BB0E4C03CD65AD@solaris.neuro.uni-luebeck.de> References: <810A8E06C75EB447A8CEB73DBFD7BB0E4C03CD65AD@solaris.neuro.uni-luebeck.de> Message-ID: Dear Nuria For the earliest components from S1, and just to get started and get a feel for your data, a traditional dipole fit using FT_DIPOLEFITTING should work fine. Following the initial localization of S1, you indeed could continue with an LCMV beamformer. There is an example at http://fieldtrip.fcdonders.nl/example/fit_a_dipole_to_the_tactile_erf_after_mechanical_stimulation The example is on CTF data with Braille-cell stimulation (i.e. mechanical instead of electrical),. Since the example data is available from the ftp you can try the example out before translating it to your own data. Furthermore I should note that the example is rather old and probably has not been run for quite some time, so there might be some pieces of the example script on the wiki that need updating to reflect the current state of the fieldtrip code. I suggest you also have a look here http://fieldtrip.fcdonders.nl/tutorial/headmodel_meg, which is a new tutorial that Lilla Magyari (CC) made available last week. best regards Robert On 21 Aug 2012, at 12:50, Nuria Doñamayor Alonso wrote: > Dear fieldtrippers, > I am currently analyzing a study, measured with a Neuromag 306 system, in which we stimulated the median nerve. I had never done source reconstruction with fieldtrip, so I followed the tutorial using MNE for EMF source analysis (which worked fine), only to find out that I could neither average individual sources nor perform stats, which is exactly what I need to do... > So the question is: What would you guys recommend? Should I just try using the LCMV beamfomer? However, that might not be the best solution according to this thread http://mailman.science.ru.nl/pipermail/fieldtrip/2011-September/004321.html... Is there any way I could efficiently export my preprocessed/averaged data to some other software (MNE/SPM/Brainstorm...) and compute my source reconstruction and stats there? What do you guys usually do? > Thanks a lot for your help, > Nuria > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From Hisako.Fujiwara at cchmc.org Tue Aug 21 22:13:04 2012 From: Hisako.Fujiwara at cchmc.org (Fujiwara, Hisako) Date: Tue, 21 Aug 2012 20:13:04 +0000 Subject: [FieldTrip] beamformer application to spontaneous signal source construction Message-ID: Dear all, I have a spontaneous recording of a patient who has epilepsy. I want to use beamformer to reconstruct and display an increase power at certain frequency and location in the brain. Your example only describe how to use beamfomer for ERP. Could you kindly advise how to apply your fieldtrip beamformer method for my needed application. I would like to thank your all kind help and advice in advance. Sincerely, Hisako -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Wed Aug 22 19:59:51 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Wed, 22 Aug 2012 19:59:51 +0200 Subject: [FieldTrip] FT_REJECTARTIFACT Message-ID: Dear all, I have one problem concerning ft_rejectartifact function. I did everything as you explained in automatic artifact rejection tutorial. I mark segments first and that is fine, but the ft_rejectartifact doesn't remove marked bad segments form my data. Please help! Here is my code: % automatic clear all s=[87 209 195 225 60]; for g=1:length(s) input_no=s(g) clear sub sub=['Subject' num2str(input_no)]; eval(sub) %calling subject's script cfg=[]; cfg.trialdef.eventtype = 'UPPT001'; cfg.trialdef.eventvalue = 1; cfg.trialdef.prestim = 0.1; cfg.trialdef.poststim = 0.5; cfg.trialfun = 'trialfun_general'; cfg.channel = {'MEG'}; cfg.continuous = 'yes'; cfg.dataset = [subjectdata.wr11.datadir]; [data1]=ft_definetrial(cfg) trial=data1.trl; % muscle cfg=[]; cfg.trl =trial; cfg.datafile = [subjectdata.wr11.datafile]; cfg.headerfile =[subjectdata.wr11.headerfile]; cfg.continuous = 'yes'; % channel selection, cutoff and padding cfg.artfctdef.zvalue.channel = 'MRT*'; cfg.artfctdef.zvalue.cutoff = 15; cfg.artfctdef.zvalue.trlpadding = 0.5; cfg.artfctdef.zvalue.fltpadding = 0.5; cfg.artfctdef.zvalue.artpadding = 0.1; % algorithmic parameters cfg.artfctdef.zvalue.bpfilter = 'yes'; cfg.artfctdef.zvalue.bpfreq = [110 140]; cfg.artfctdef.zvalue.bpfiltord = 9; cfg.artfctdef.zvalue.bpfilttype = 'but'; cfg.artfctdef.zvalue.hilbert = 'yes'; cfg.artfctdef.zvalue.boxcar = 0.2; %make the process interactive cfg.artfctdef.zvalue.interactive = 'yes'; [cfg, artifact_muscle] = ft_artifact_zvalue(cfg) % jump cfg=[]; cfg.trl = trial; cfg.datafile = [subjectdata.wr11.datafile]; cfg.headerfile =[subjectdata.wr11.headerfile]; cfg.continuous = 'yes'; % channel selection, cutoff and padding cfg.artfctdef.zvalue.channel = 'MEG'; cfg.artfctdef.zvalue.cutoff = 20; cfg.artfctdef.zvalue.trlpadding = 0; cfg.artfctdef.zvalue.artpadding = 0; cfg.artfctdef.zvalue.fltpadding = 0; % algorithmic parameters cfg.artfctdef.zvalue.cumulative = 'yes'; cfg.artfctdef.zvalue.medianfilter = 'yes'; cfg.artfctdef.zvalue.medianfiltord = 9; cfg.artfctdef.zvalue.absdiff = 'yes'; % make the process interactive cfg.artfctdef.zvalue.interactive = 'yes'; [cfg, artifact_jump] = ft_artifact_zvalue(cfg) cfg.artfctdef.reject = 'complete'; cfg.artfctdef.jump.artifact = artifact_jump; cfg.artfctdef.muscle.artifact = artifact_muscle; data_no_artifacts = ft_rejectartifact(data1) end Thank you in advance! NP -------------- next part -------------- An HTML attachment was scrubbed... URL: From eelke.spaak at donders.ru.nl Wed Aug 22 20:17:29 2012 From: eelke.spaak at donders.ru.nl (Eelke Spaak) Date: Wed, 22 Aug 2012 20:17:29 +0200 Subject: [FieldTrip] FT_REJECTARTIFACT In-Reply-To: References: Message-ID: Dear Nenad, My guess is you are missing the first argument in your call to ft_rejectartifact; i.e. instead of data_no_artifacts = ft_rejectartifact(data1) you should use data_no_artifacts = ft_rejectartifact(cfg, data1) Does that work? Best, Eelke On 22 August 2012 19:59, Nenad Polomac wrote: > Dear all, > > I have one problem concerning ft_rejectartifact function. > I did everything as you explained in automatic artifact rejection tutorial. > I mark segments first and that is fine, but the ft_rejectartifact doesn't > remove marked bad segments form my data. > Please help! > > Here is my code: > > % automatic > clear all > > s=[87 209 195 225 60]; > > > for g=1:length(s) > input_no=s(g) > > clear sub > > sub=['Subject' num2str(input_no)]; > eval(sub) %calling subject's script > > > cfg=[]; > cfg.trialdef.eventtype = 'UPPT001'; > cfg.trialdef.eventvalue = 1; > cfg.trialdef.prestim = 0.1; > cfg.trialdef.poststim = 0.5; > cfg.trialfun = 'trialfun_general'; > cfg.channel = {'MEG'}; > cfg.continuous = 'yes'; > cfg.dataset = [subjectdata.wr11.datadir]; > > [data1]=ft_definetrial(cfg) > trial=data1.trl; > > > % muscle > cfg=[]; > cfg.trl =trial; > cfg.datafile = [subjectdata.wr11.datafile]; > cfg.headerfile =[subjectdata.wr11.headerfile]; > cfg.continuous = 'yes'; > > % channel selection, cutoff and padding > cfg.artfctdef.zvalue.channel = 'MRT*'; > cfg.artfctdef.zvalue.cutoff = 15; > cfg.artfctdef.zvalue.trlpadding = 0.5; > cfg.artfctdef.zvalue.fltpadding = 0.5; > cfg.artfctdef.zvalue.artpadding = 0.1; > > % algorithmic parameters > cfg.artfctdef.zvalue.bpfilter = 'yes'; > cfg.artfctdef.zvalue.bpfreq = [110 140]; > cfg.artfctdef.zvalue.bpfiltord = 9; > cfg.artfctdef.zvalue.bpfilttype = 'but'; > cfg.artfctdef.zvalue.hilbert = 'yes'; > cfg.artfctdef.zvalue.boxcar = 0.2; > > %make the process interactive > cfg.artfctdef.zvalue.interactive = 'yes'; > > [cfg, artifact_muscle] = ft_artifact_zvalue(cfg) > > > > % jump > cfg=[]; > cfg.trl = trial; > cfg.datafile = [subjectdata.wr11.datafile]; > cfg.headerfile =[subjectdata.wr11.headerfile]; > cfg.continuous = 'yes'; > > % channel selection, cutoff and padding > cfg.artfctdef.zvalue.channel = 'MEG'; > cfg.artfctdef.zvalue.cutoff = 20; > cfg.artfctdef.zvalue.trlpadding = 0; > cfg.artfctdef.zvalue.artpadding = 0; > cfg.artfctdef.zvalue.fltpadding = 0; > > % algorithmic parameters > cfg.artfctdef.zvalue.cumulative = 'yes'; > cfg.artfctdef.zvalue.medianfilter = 'yes'; > cfg.artfctdef.zvalue.medianfiltord = 9; > cfg.artfctdef.zvalue.absdiff = 'yes'; > > % make the process interactive > cfg.artfctdef.zvalue.interactive = 'yes'; > > [cfg, artifact_jump] = ft_artifact_zvalue(cfg) > > > > > cfg.artfctdef.reject = 'complete'; > cfg.artfctdef.jump.artifact = artifact_jump; > cfg.artfctdef.muscle.artifact = artifact_muscle; > data_no_artifacts = ft_rejectartifact(data1) > end > > Thank you in advance! > NP > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From polomacnenad at gmail.com Thu Aug 23 10:50:20 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Thu, 23 Aug 2012 10:50:20 +0200 Subject: [FieldTrip] FT_REJECTARTIFACT Message-ID: Dear Eelke, Thank you for your suggestion. But unfortunately this doesn't work. If I do your way I get: Error using ft_rejectartifact (line 233) no trials were selected, cannot perform artifact detection/rejection. So I am still not sure what might be the problem... Regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From stan.vanpelt at fcdonders.ru.nl Thu Aug 23 11:25:21 2012 From: stan.vanpelt at fcdonders.ru.nl (Stan van Pelt) Date: Thu, 23 Aug 2012 11:25:21 +0200 (CEST) Subject: [FieldTrip] FT_REJECTARTIFACT In-Reply-To: Message-ID: <1052918099.53046.1345713921693.JavaMail.root@sculptor.zimbra.ru.nl> Dear Nenad, It seems that no trials are defined when you run ft_definetrial. Can you check if there are actually is trial data present in your data1 structure? If there is no data, you should check if your configuration settings are correct in that processing step (e.g., trigger channel, event values). Best, Stan ----- Oorspronkelijk bericht ----- > Van: "Nenad Polomac" > Aan: fieldtrip at science.ru.nl > Verzonden: Donderdag 23 augustus 2012 10:50:20 > Onderwerp: Re: [FieldTrip] FT_REJECTARTIFACT > Dear Eelke, > Thank you for your suggestion. But unfortunately this doesn't work. If > I do your way I get: Error using ft_rejectartifact (line 233) > no trials were selected, cannot perform artifact detection/rejection. > So I am still not sure what might be the problem... > Regards! > Nenad > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Stan van Pelt Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen Kapittelweg 29, 6525 EN Nijmegen, Netherlands E-mail: stan.vanpelt at donders.ru.nl Website: www.ru.nl/donders/ Phone: (+31) (0)24 36 10981 Fax: (+31) (0)24 36 10989 -------------- next part -------------- An HTML attachment was scrubbed... URL: From poortjasper at gmail.com Thu Aug 23 12:34:48 2012 From: poortjasper at gmail.com (Jasper Poort) Date: Thu, 23 Aug 2012 11:34:48 +0100 Subject: [FieldTrip] Fwd: ft_freqdescriptives and cfg.trial In-Reply-To: References: Message-ID: Dear Fieldtrip users, I have a problem with using a trial selection to do freqdescriptives on a subset of trials: I first do ft_freqanalysis with these setttings, which results in freq: cfg.output = 'fourier'; cfg.method = 'mtmconvol'; cfg.keeptrials = 'yes'; freq = label: {1x64 cell} dimord: 'rpttap_chan_freq_time' freq: [30 35 40 45 50 55 60 65 70 75 80 85 90 95 100] time: [1x36 double] fourierspctrm: [4-D double] cumtapcnt: [857x15 double] cfg: [1x1 struct] size(freq.fourierspctrm) ans = 2571 64 15 36 then i try running ft_freqdescriptives with these settings tmpcfg = []; tmpcfg.jackknife = 'no'; tmpcfg.keeptrials = 'no'; tmpcfg.channel = freq.label([chnix1,chnix2]) tmpcfg.trials = [1:10:100]; tmp = ft_freqdescriptives(tmpcfg,freq); Error using + Array dimensions must match for binary array op. Error in ft_checkdata>fixcsd (line 740) powspctrm = powspctrm + abs(data.fourierspctrm(p:ntap:end,:,:,:,:)).^2; Error in ft_checkdata (line 646) data = fixcsd(data, cmbrepresentation, channelcmb); Error in ft_freqdescriptives (line 131) freq = ft_checkdata(freq, 'cmbrepresentation', 'sparsewithpow', 'channelcmb', {}); the error seems to happen because the cumtapcnt seems to be updated after calling ft_selectdata in ft_freqdescriptives such that it no longer contains a row for every trial which causes nrpt = size(data.cumtapcnt,1); in ft_checkdata to return 1 instead of the number of trials. label: {2x1 cell} dimord: 'rpttap_chan_freq_time' freq: [30 35 40 45 50 55 60 65 70 75 80 85 90 95 100] time: [1x36 double] fourierspctrm: [4-D double] cumtapcnt: [3 3 3 3 3 3 3 3 3 3] cfg: [1x1 struct] I was wondering if this is a bug or am I somehow specifying the input incorrectly? Thanks! best, Jasper -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 23 12:55:02 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 23 Aug 2012 12:55:02 +0200 Subject: [FieldTrip] Fwd: ft_freqdescriptives and cfg.trial In-Reply-To: References: Message-ID: Hi Jasper, I think this is indeed a bug in fieldtrip. The reason being that the cumtapcnt field for timefrequency data is not well-defined/well-designed/well-handled (the latter by ft_selectdata, which is doing the trial selection). Probably you can work around this by updating the cumtapcnt field prior to calling ft_freqdescriptives. I suspect that it will work if you define it to be a vector (nx1) where n is the number of trials, and each entry has the value of the number of tapers applied (per trial). Would it be possible for you to create a bug on our bugzilla site (bugzilla.fcdonders.nl) with a description of the problem, a little test script and some (small) data for reproduction? The team will then take it from there. Best, Jan-Mathijs On Aug 23, 2012, at 12:34 PM, Jasper Poort wrote: > Dear Fieldtrip users, > > > > I have a problem with using a trial selection to do freqdescriptives on a subset of trials: > > > > I first do ft_freqanalysis with these setttings, which results in freq: > > > > cfg.output = 'fourier'; > > cfg.method = 'mtmconvol'; > > cfg.keeptrials = 'yes'; > > > freq = > > > label: {1x64 cell} > > dimord: 'rpttap_chan_freq_time' > > freq: [30 35 40 45 50 55 60 65 70 75 80 85 90 95 100] > > time: [1x36 double] > > fourierspctrm: [4-D double] > > cumtapcnt: [857x15 double] > > cfg: [1x1 struct] > > > size(freq.fourierspctrm) > > ans = > > 2571 64 15 36 > > > then i try running ft_freqdescriptives with these settings > > tmpcfg = []; > > tmpcfg.jackknife = 'no'; > > tmpcfg.keeptrials = 'no'; > > tmpcfg.channel = freq.label([chnix1,chnix2]) > > tmpcfg.trials = [1:10:100]; > > tmp = ft_freqdescriptives(tmpcfg,freq); > > > Error using + > > Array dimensions must match for binary array op. > > > Error in ft_checkdata>fixcsd (line 740) > > powspctrm = powspctrm + abs(data.fourierspctrm(p:ntap:end,:,:,:,:)).^2; > > > Error in ft_checkdata (line 646) > > data = fixcsd(data, cmbrepresentation, channelcmb); > > > Error in ft_freqdescriptives (line 131) > > freq = ft_checkdata(freq, 'cmbrepresentation', 'sparsewithpow', 'channelcmb', {}); > > > > the error seems to happen because the cumtapcnt seems to be updated after calling ft_selectdata in ft_freqdescriptives such that it no longer contains a row for every trial which causes nrpt = size(data.cumtapcnt,1); in ft_checkdata to return 1 instead of the number of trials. > > > label: {2x1 cell} > dimord: 'rpttap_chan_freq_time' > freq: [30 35 40 45 50 55 60 65 70 75 80 85 90 95 100] > time: [1x36 double] > fourierspctrm: [4-D double] > cumtapcnt: [3 3 3 3 3 3 3 3 3 3] > cfg: [1x1 struct] > > > > I was wondering if this is a bug or am I somehow specifying the input incorrectly? > > > > Thanks! best, Jasper > > > > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jm.horschig at donders.ru.nl Thu Aug 23 13:41:00 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Thu, 23 Aug 2012 13:41:00 +0200 Subject: [FieldTrip] FT_REJECTARTIFACT In-Reply-To: References: Message-ID: <503616CC.7010303@donders.ru.nl> Dear Nenad, it is a bit confusing that you call the variable returned from ft_definetrial 'data1', because actually it is only a cfg-structure without data. No matter how you call that variable though, it should work if you use this: / data1.artfctdef.reject = 'complete'; / / data1.artfctdef.jump.artifact = artifact_jump;/ / data1.artfctdef.muscle.artifact = artifact_muscle;/ / data_no_artifacts = ft_rejectartifact(data1)/ Note that you need to call ft_preprocessing in order to obtain data, in your case: /data = ft_preprocessing(data_no_artifacts);/ or /data = ft_preprocessing(data1);/ ft_definetrial is a function which just returns at what samples your trials are supposed to start. ft_preprocessing is then using this information to extract the trials out of your datafile. You can artifact rejection either before you segmented the data into trials, as done here, or afterwards, i.e. after having called ft_preprocessing. Best, Jörn On 8/23/2012 10:50 AM, Nenad Polomac wrote: > Dear Eelke, > > Thank you for your suggestion. But unfortunately this doesn't work. If > I do your way I get: Error using ft_rejectartifact (line 233) > no trials were selected, cannot perform artifact detection/rejection. > > So I am still not sure what might be the problem... > > Regards! > > Nenad > > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group FieldTrip Development Team P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Thu Aug 23 14:46:37 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Thu, 23 Aug 2012 14:46:37 +0200 Subject: [FieldTrip] FT_REJECTARTIFACT In-Reply-To: References: Message-ID: Dear Jörn, Thank you for your answer! That was the problem. I understand all now. Regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From s.rombetto at cib.na.cnr.it Thu Aug 23 14:57:51 2012 From: s.rombetto at cib.na.cnr.it (s.rombetto at cib.na.cnr.it) Date: Thu, 23 Aug 2012 14:57:51 +0200 Subject: [FieldTrip] EDF data Message-ID: <20120823145751.avcc2gkcw0ws0wow@arco.cib.na.cnr.it> Dear fieldtrip users, I'm trying to import some EEG data in EDF format. I am able to read the data structure using hdr= ft_read_header(rawdataname); and also to read the data by data = ft_read_data(rawdataname);, but I receive an error message when I try to perform the ft_preprocessing command (Error in ==> ft_preprocessing at 511). Do you have any hint or sample code? Thank you in advance! Sara ------------------------- Dott.ssa Sara Rombetto Istituto di Cibernetica "E. Caianiello" Via Campi Flegrei, 34 80078 Pozzuoli (NA) Italy tel +390818675361 fax +390818675128 -------------------------- "I disapprove of what you say, but I will defend to the death your right to say it." [Evelyn Beatrice Hall, The Friends Of Voltaire] ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From nuria.donamayor at neuro.uni-luebeck.de Thu Aug 23 16:03:04 2012 From: nuria.donamayor at neuro.uni-luebeck.de (=?iso-8859-1?Q?Nuria_Do=F1amayor_Alonso?=) Date: Thu, 23 Aug 2012 16:03:04 +0200 Subject: [FieldTrip] source reconstruction on emfs In-Reply-To: References: <810A8E06C75EB447A8CEB73DBFD7BB0E4C03CD65AD@solaris.neuro.uni-luebeck.de>, Message-ID: <810A8E06C75EB447A8CEB73DBFD7BB0E4C03CD65C1@solaris.neuro.uni-luebeck.de> Dear Robert, I did what you suggested and everything worked perfectly, thanks. Now I have a slight problem with the following. I want to delimit a ROI to do stats on my LCMV results and, as far as I understand, I should use ft_sourceinterpolate, ft_volumenormalise and then ft_sourcegrandaverage. The problem is, when I try to check out the result with ft_sourceplot, it just plots the first slice 20 times. This does not happen with the non-normalised grandaverages (and the individual averages in both cases). Any suggestions? Here's the code I'm using: # on the individual data cfg = []; cdf.downsample = 2; cfg.parameter = 'avg.nai'; sourceint = ft_sourceinterpolate(cfg, source, mri); cfg = []; cfg.coordsys = 'neuromag'; cfg.nonlinear = 'no'; sourcenorm = ft_volumenormalise(cfg, sourceint); # grandaverage cfg = []; cfg.keepindividual = 'yes'; cfg.parameter = 'nai'; gsource = (cfg, sourceint1, sourceint2...); # plot gsource cfg = []; cfg.funparameter = 'avg.nai'; cfg.maskparameter = cfg.funparameter; ft_sourceplot(cfg, gsource); Thanks again! Nuria ________________________________________ Von: fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl] im Auftrag von Robert Oostenveld [r.oostenveld at donders.ru.nl] Gesendet: Dienstag, 21. August 2012 21:56 An: FieldTrip discussion list Betreff: Re: [FieldTrip] source reconstruction on emfs Dear Nuria For the earliest components from S1, and just to get started and get a feel for your data, a traditional dipole fit using FT_DIPOLEFITTING should work fine. Following the initial localization of S1, you indeed could continue with an LCMV beamformer. There is an example at http://fieldtrip.fcdonders.nl/example/fit_a_dipole_to_the_tactile_erf_after_mechanical_stimulation The example is on CTF data with Braille-cell stimulation (i.e. mechanical instead of electrical),. Since the example data is available from the ftp you can try the example out before translating it to your own data. Furthermore I should note that the example is rather old and probably has not been run for quite some time, so there might be some pieces of the example script on the wiki that need updating to reflect the current state of the fieldtrip code. I suggest you also have a look here http://fieldtrip.fcdonders.nl/tutorial/headmodel_meg, which is a new tutorial that Lilla Magyari (CC) made available last week. best regards Robert On 21 Aug 2012, at 12:50, Nuria Doñamayor Alonso wrote: > Dear fieldtrippers, > I am currently analyzing a study, measured with a Neuromag 306 system, in which we stimulated the median nerve. I had never done source reconstruction with fieldtrip, so I followed the tutorial using MNE for EMF source analysis (which worked fine), only to find out that I could neither average individual sources nor perform stats, which is exactly what I need to do... > So the question is: What would you guys recommend? Should I just try using the LCMV beamfomer? However, that might not be the best solution according to this thread http://mailman.science.ru.nl/pipermail/fieldtrip/2011-September/004321.html... Is there any way I could efficiently export my preprocessed/averaged data to some other software (MNE/SPM/Brainstorm...) and compute my source reconstruction and stats there? What do you guys usually do? > Thanks a lot for your help, > Nuria > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip _______________________________________________ fieldtrip mailing list fieldtrip at donders.ru.nl http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From kamx at ymail.com Fri Aug 24 06:56:53 2012 From: kamx at ymail.com (Kam X) Date: Thu, 23 Aug 2012 21:56:53 -0700 (PDT) Subject: [FieldTrip] Problem setting up fieldtrip Message-ID: <1345784213.64205.YahooMailClassic@web162305.mail.bf1.yahoo.com> Hi, I hope this is the proper type of questions for this list, if not I apologize.I've just set up fieldtrip by unzipping it, adding the path to the folder to matlab, and running ft_defaults.  Most things seem to work, but if I try to use the command ft_realtime_signalproxy I get: Undefined function 'ft_realtime_signalproxy' for input arguments of type 'struct'. Looking in the fieldtrip folder, I see the .m files, but ft_realtime_signalproxy is not among them.  If I run realtime_signalproxy(cfg), I get: Warning: REALTIME_SIGNALPROXY is only a compatibility wrapper, which will soon be removed.Please instead call FT_REALTIME_SIGNALPROXY. > In realtime_signalproxy at 16 Undefined function 'ft_realtime_signalproxy' for input arguments of type 'struct'. Error in realtime_signalproxy (line 18)[varargout{1:nargout}] = funhandle(varargin{:}); Have I missed a step in setting up fieldtrip, or do I need to download the realtime functions from somewhere? Thanks,Kameron -------------- next part -------------- An HTML attachment was scrubbed... URL: From jm.horschig at donders.ru.nl Fri Aug 24 09:51:06 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Fri, 24 Aug 2012 09:51:06 +0200 Subject: [FieldTrip] Problem setting up fieldtrip In-Reply-To: <1345784213.64205.YahooMailClassic@web162305.mail.bf1.yahoo.com> References: <1345784213.64205.YahooMailClassic@web162305.mail.bf1.yahoo.com> Message-ID: <5037326A.8030209@donders.ru.nl> Dear Kameron, Thanks for reporting this. FieldTrip is modularly organized, with 'realtime' being a separate module/toolbox. Usually, modules should be added automatically when calling ft_defaults. The ft_hastoolbox function should take care of this. However, the realtime folder has been restructured recently so that it has subfolders now, which are (unfortunately) not added by default. It is a(tiny little) bit more tedious now to add this to your path manually: /ft_hastoolbox('realtime/example', 1)/ The same holds for any other subfolders of the realtime module. I'm gonna check with Robert (probably after the BioMag conference) how to resolve this best (if at all). Best, Jörn On 8/24/2012 6:56 AM, Kam X wrote: > Hi, > > I hope this is the proper type of questions for this list, if not I > apologize. > I've just set up fieldtrip by unzipping it, adding the path to the > folder to matlab, and running ft_defaults. Most things seem to work, > but if I try to use the command ft_realtime_signalproxy I get: > > Undefined function 'ft_realtime_signalproxy' for input arguments of > type 'struct'. > > Looking in the fieldtrip folder, I see the .m files, but > ft_realtime_signalproxy is not among them. If I run > realtime_signalproxy(cfg), I get: > > Warning: REALTIME_SIGNALPROXY is only a compatibility wrapper, which > will soon be removed. > Please instead call FT_REALTIME_SIGNALPROXY. > > In realtime_signalproxy at 16 > Undefined function 'ft_realtime_signalproxy' for input arguments of > type 'struct'. > > Error in realtime_signalproxy (line 18) > [varargout{1:nargout}] = funhandle(varargin{:}); > > Have I missed a step in setting up fieldtrip, or do I need to download > the realtime functions from somewhere? > > Thanks, > Kameron > > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group FieldTrip Development Team P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From bobrobbrown at googlemail.com Fri Aug 24 10:38:58 2012 From: bobrobbrown at googlemail.com (Robert Brown) Date: Fri, 24 Aug 2012 11:38:58 +0300 Subject: [FieldTrip] PLV on sources Message-ID: Dear Fieldtrippers, I see (from the ft_connectivityanalysis code) that one cannot run phase locking (PLV) analysis (a la Lachaux et al) on source data. My brain still insisted on trying to do so but unfortunately I am not sure if it is clever enough. Hope you can help! It seems that a way to go would be to do LCMV beamforming on the data, get the virtual electrodes, then run freqanalysis with fourier output on these virtual electrodes and then perform PLV on those data. Before I dig myself very deep into this, maybe some experts here could guide me: Would this approach work? Any possible caveats? Are there maybe other, better and more straightforward ways for achieving the PLV on source level? Any comments would be appreciated. Thank you very much for your time! Kind regards, Bob -------------- next part -------------- An HTML attachment was scrubbed... URL: From s.bogels at psy.gla.ac.uk Fri Aug 24 13:25:48 2012 From: s.bogels at psy.gla.ac.uk (Sara =?iso-8859-1?b?QvZnZWxz?=) Date: Fri, 24 Aug 2012 12:25:48 +0100 Subject: [FieldTrip] How to specify two dipoles in cfg.supdip in ft_sourceanalysis? Message-ID: <20120824122548.739016jhmbe93sn0@horde.psy.gla.ac.uk> Hi all, I am using ft_sourceanalysis with specifications of cfg.refdip and cfg.supdip. Until now, I always inserted the same voxel coordinates in supdip as refdip to suppress self-coherence. However, the help suggests that you can also suppress more than one dipole, but it does not specify how to input this in cfg.supdip. I tried a 2 by 3 matrix and a cell structure with two entries (1 x 3 matrices) but that gave (apparently) the same result as suppressing only refdip. As a related, more general question, what about the orientation of the (e.g. reference) dipole? How is that actually determined in this case (since you only have to insert the position of the dipole)? Thank you, Sara ---------------------------------------------------------------- This message was sent using the Web mail system for The University of Glasgow School of Psychology ------------------------------------------------------------------ From narayan.ps at tut.fi Fri Aug 24 14:35:40 2012 From: narayan.ps at tut.fi (Narayan Puthanmadam Subramaniyam) Date: Fri, 24 Aug 2012 12:35:40 +0000 Subject: [FieldTrip] PLV on sources Message-ID: <201208241235.q7OCZhaZ002930@brievenbus.science.ru.nl> hi i was thinking of doing the same thing exactly. Computing plv on the eeg inverse solution. technically it should be possible going by the general level concept of phase locking. but i am not sure so let the experts here speak. Thanks&Regards, NPS Sent from my Nokia phone -----Original Message----- From: Robert Brown Sent: 24:08:2012, 11:38 To: fieldtrip at science.ru.nl Subject: [FieldTrip] PLV on sources Dear Fieldtrippers, I see (from the ft_connectivityanalysis code) that one cannot run phase locking (PLV) analysis (a la Lachaux et al) on source data. My brain still insisted on trying to do so but unfortunately I am not sure if it is clever enough. Hope you can help! It seems that a way to go would be to do LCMV beamforming on the data, get the virtual electrodes, then run freqanalysis with fourier output on these virtual electrodes and then perform PLV on those data. Before I dig myself very deep into this, maybe some experts here could guide me: Would this approach work? Any possible caveats? Are there maybe other, better and more straightforward ways for achieving the PLV on source level? Any comments would be appreciated. Thank you very much for your time! Kind regards, Bob _______________________________________________ fieldtrip mailing list fieldtrip at donders.ru.nl http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From maximilien.chaumon at gmail.com Fri Aug 24 17:13:22 2012 From: maximilien.chaumon at gmail.com (Maximilien Chaumon) Date: Fri, 24 Aug 2012 17:13:22 +0200 Subject: [FieldTrip] [Eeglablist] nested hypothesis testing to decide whether to use one or two dipoles to fit a component In-Reply-To: References: Message-ID: Hi everyone, Just to follow-up on this if anyone wanted to take that approach. I did use the following procedure to decide whether to fit one or two dipoles to a given component. 1) fit one dipole, compute SSE for each component map as the RV * sum of squared inverse weights (sse(1) = EEG.dipfit.model(idip).rv .* sum(EEG.icawinv(:,idip).^2)) 2) fit two dipoles, compute SSE for each component map: sse(2) 3) assume p1 = 6 parameters estimated for 1 dipole, p2 = 9 parameters estimated for 2 dipoles (only three momentum values more, since symmetry imposes position for the two dipoles), n = EEG.nbchan 4) F statistic : F = (sse(1)-sse(2))/(p2-p1)/(sse(2)/(n-p2)) 5) if F > finv(.95,p2-p1,n-p2) decide to go for 2 dipoles. 6) that yielded some solutions where 2 dipoles were selected but were in fact at the same position (x == 0) with opposite momenta. Some others had really just one dipole with large amplitude and the other much smaller. So I discarded solutions where distance between the two dipoles was less than 5mm, or where the momentum angles were opposed, or when the amplitude ratio between the two dipoles was bigger than 5. The final solution is only moderately satisfying. Looking at the maps, the 2 dipole solution is still too often favored over the one dipole solution. As Scott previously mentioned, it could be that the second dipole accounts for error in the head model, and that's all. The F difference in SSE is probably not a good measure of the two-dipolarity of a component. One would probably better just screen all components by eye and spot those that look dipolar using thumbometric visual assessment. Best, Max PS: the script I used: dofit = 1; dorework = 0; doplot = 1; rootdir = '/DATA/'; cd(rootdir) studyfile = fullfile(rootdir,'AP2.study'); if dofit || dorework [STUDY ALLEEG] = pop_loadstudy(studyfile); for i_set = 1:numel(STUDY.datasetinfo) EEG = pop_loadset('filename',ALLEEG(i_set).filename,'filepath',ALLEEG(i_set).filepath);%pop_loadset(GRAB(i_set).name); if isfield(EEG,'peakalphafrequency') EEG.peakalphafreq = EEG.peakalphafrequency; EEG = rmfield(EEG,'peakalphafrequency'); end if ~dorework EEG = pop_dipfit_settings( EEG, 'hdmfile',[ fileparts(which('eeglab')) '/plugins/dipfit2.2/standard_BEM/standard_vol.mat'],'coordformat','MNI','mrifile',[ fileparts(which('eeglab')) '/plugins/dipfit2.2/standard_BEM/standard_mri.mat'],'chanfile',[ fileparts(which('eeglab')) '/plugins/dipfit2.2/standard_BEM/elec/standard_1005.elc'],'coord_transform',[0.67468 -15.6307 2.426 0.081417 0.0024349 -1.5728 1.1744 1.0622 1.1487] ,'chansel',[1:EEG.nbchan] ); EEGonedipole = pop_multifit(EEG, [1:size(EEG.icawinv,2)] ,'threshold',100,'rmout','on','dipoles',1); EEGtwodipoles = pop_multifit(EEG, [1:size(EEG.icawinv,2)] ,'threshold',100,'rmout','on','dipoles',2); else EEGonedipole = EEG; EEGtwodipoles = EEG; EEGonedipole.dipfit.model = EEG.dipfit.onedipole; EEGtwodipoles.dipfit.model = EEG.dipfit.twodipoles; end % H0: one dipole. EEG.dipfit = EEGonedipole.dipfit; % still we store them both. doesn't hurt. EEG.dipfit.onedipole = EEGonedipole.dipfit.model; EEG.dipfit.twodipoles = EEGtwodipoles.dipfit.model; % now for every component, if the rv is much better for the % 2dipoles than for the 1dipole, we'll use the 2dipoles. for i_dip = 1:numel(EEGonedipole.dipfit.model) sse(1) = EEGonedipole.dipfit.model(i_dip).rv .* sum(EEGonedipole.icawinv(:,i_dip).^2); sse(2) = EEGtwodipoles.dipfit.model(i_dip).rv .* sum(EEGtwodipoles.icawinv(:,i_dip).^2); p1 = 6;% six params for 1dipole model p2 = 9;% 9 for 2 dipoles model (assume only symmetry is fixed). n = EEG.nbchan; % N is number of electrodes % F statistic F = (sse(1)-sse(2))/(p2-p1)/(sse(2)/(n-p2)); EEG.dipfit.Fone2two(i_dip) = F; % F decision Fthresh = finv(.999,p2-p1,n-p2); if F > Fthresh % reject H0, take 2 dipoles model % only if x ~= 0 and the two momenta are not just opposite. mom = EEGtwodipoles.dipfit.model(i_dip).momxyz; pos = EEGtwodipoles.dipfit.model(i_dip).posxyz; [t p r] = cart2sph(mom(:,1),mom(:,2),mom(:,3)); t(2) = t(2)+pi;% d = sqrt(sum(diff(pos).^2)); % if the two dipoles are too close and have exactly % opposite angles cond = d < 5 && (diff(abs(t)) < 0.01 || diff(abs(p)) < 0.01); % or if they are exactly on the sagittal plane cond = cond | abs(EEGtwodipoles.dipfit.model(i_dip).posxyz(1)) < 5; % or if the norm of one is much bigger than the other ratio = norm(mom(1,:)) ./ norm(mom(2,:)); ratiothresh = 5; cond = cond | (ratio > ratiothresh || ratio < 1/ratiothresh); % we will still keep the one dipole fit if ~cond% so if not, we take the 2 dipole fit. EEG.dipfit.model(i_dip) = EEGtwodipoles.dipfit.model(i_dip); end end end % reject poor overall fits. EEG.dipfit.model = dipfit_reject(EEG.dipfit.model, 20/100); % reject NaN rv for i_m = 1:numel(EEG.dipfit.model) if isnan(EEG.dipfit.model(i_m).rv) EEG.dipfit.model(i_m).posxyz = []; EEG.dipfit.model(i_m).momxyz = []; EEG.dipfit.model(i_m).rv = 1; end end % removing dipoles outside the head % --------------------------------- rmdip = []; for index = 1:numel(EEGonedipole.dipfit.model) if ~isempty(EEG.dipfit.model(index).posxyz) if any(sqrt(sum(EEG.dipfit.model(index).posxyz.^2,2)) > 85) rmdip = [ rmdip index]; EEG.dipfit.model(index).posxyz = []; EEG.dipfit.model(index).momxyz = []; EEG.dipfit.model(index).rv = 1; end; end; end; if ~isempty(rmdip) fprintf('%d out of cortex dipoles removed (usually artifacts)\n', length(rmdip)); end; pop_saveset(EEG,'savemode','resave'); end end %% if doplot % [STUDY ALLEEG] = pop_loadstudy('AlphaPsyc2.study'); num = [];% number of dipoles that are fit for each component idx = []; topos = {[] [] []}; Fdiff = {[] [] []}; idx = {[] [] []}; dips = {}; for i_e = 1:numel(ALLEEG) for i_comp = 1:numel(ALLEEG(i_e).dipfit.model) num(end+1) = size(ALLEEG(i_e).dipfit.model(i_comp).posxyz,1); topos{num(end)+1}(end+1,:) = ALLEEG(i_e).icawinv(:,i_comp)'; dips{num(end)+1}(size(topos{num(end)+1},1)) = ALLEEG(i_e).dipfit.model(i_comp); Fdiff{num(end)+1}(end+1) = ALLEEG(i_e).dipfit.Fone2two(i_comp); idx{num(end)+1}(end+1,:) = [i_e i_comp]; end end for i_plot = 2:3 figure(6548+i_plot);paste_figpos manytopos ntopos = size(topos{i_plot},1); n = ceil(sqrt(ntopos)); m = n; for i_topo = 1:ntopos subplot(n,m,i_topo) cla topoplot(topos{i_plot}(i_topo,:),ALLEEG(1).chanlocs,'conv','on','electrodes','off'); title({num2str(idx{i_plot}(i_topo,:)) num2str(Fdiff{i_plot}(i_topo))}); drawnow end end end 2012/8/13 Maximilien Chaumon > Hello eeglab & Fieldtrip, > > I'm trying to find out if it would be possible to use a nested hypothesis > testing approach to decide whether to use a one or two dipole model while > estimating components' dipole locations. > The rationale I would like to follow is this: with two dipoles, we will > always obtain a better fit than with one dipole, but the decrease in sum of > squared errors (SSE) should follow a F distribution with k (= > Nparameters_2dipoles - Nparameters_1dipole) degrees of freedom. If the > decrease in SSE is greater than what would be expected under this F > distribution, then we decide that 2 dipoles provide a sufficiently better > fit and decide using them. > > I asked this question to eeglablist and Scott pointed out that it is > difficult/impossible(?) to determine if the second dipole fits actual > interesting data or just noise introduced by the imperfect head model. > Christian then said it'd be worth a shot, and I agree, so here I am again > with two questions, or two confirmations: > > 1) *How many parameters are estimated in ft_dipolefitting.m ?* specially > in the case of 2 dipoles. If I count correctly, we estimate 6 parameters > for one dipole, and, depending on whether the orientation has to be the > same in the 2 dipoles, one (amplitude) or three (amplitude and orientation) > more. > 2) *Can I assimilate the relative residual variance to a SSE?* the > function rv.m does this: rv = sum((d1-d2).^2) ./ sum(d1.^2); > So that seems to be a sum of squared errors divided by the variance of the > original data. So if I multiply the rv by the sum squared component map, I > should get it, right? > > Thanks a lot! > Max > > > 2012/8/11 Christian Kothe > >> I can only speak from my armchair here, but it sounds like it should be >> worth a try - even if you don't get the # of parameters exactly right it >> will probably give you at least some level of complexity control in >> whatever the range of validity is. If it works, it may inspire follow-up >> work (e.g., Bayesian model selection or likelihood ratio tests). >> >> The number of parameters for a 2-dipole model seems to be 3 (xyz) + 4 >> (2x the orientation parameters). Not sure about the momentum, though - you >> might look up the place where the actual function minimization is being >> performed in dipfit (fminunc call?) and see whether these are being >> optimized together with the others. >> >> Christian >> >> On Aug 10, 2012, at 3:29 PM, Scott Makeig wrote: >> >> MAX - Unfortunately, in general using two dipoles rather than one will >> ~always improve the fit. Even if the source is a pure single dipole, a >> second dipole can be used to correct for noise or errors in the forward >> head model. This is less always the case for the constrained >> spatially-symmetric dipole pairs allowed by dipfit(). However, we have not >> thought of an optimal way to decide between using one or (occasionally) two >> dipoles to fit e.g. maps of ICA brain sources. The goal would be to decide >> whether the two-dipole version is fitting noise/forward model error vs >> actual bilateral source generation... >> >> Scott Makeig >> >> On Thu, Aug 9, 2012 at 1:54 AM, Maximilien Chaumon < >> maximilien.chaumon at gmail.com> wrote: >> >>> Hello all, >>> >>> When fitting dipoles to components, we are all sooner or later puzzled >>> by the question whether to use one or two symmetrical dipoles. >>> >>> Would it be correct to put the problems in terms of a nested hypothesis >>> testing? >>> >>> We are fitting a scalp map with one or two parameters and get a residual >>> variance after the fit. >>> Could we not use this residual variance as a measure of the SSE and >>> compute a F statistic to decide whether to use the more complex (with two >>> dipoles) or simpler (with one dipole) of two nested models? >>> If yes, then how would we decide on the number of degrees of freedom? >>> How many free parameters do we have in each case? x,y,z,and two >>> orientations per dipole? how does the imposed symmetry affect that number? >>> Could we really map residual variance to SSE? How many "observations" do we >>> have in that case (see formula below)? >>> >>> I found this formula, for F: >>> F = (SSEF-SSER)/ (kF-kR) / ((1-SSEF)/(N-kF-1)) >>> where >>> SSE is sum of squared errors, >>> k is numbers of parameters, >>> N is number of observations (? what in our case?) >>> F and R indices for full and reduced model respectively (in our case two >>> and one dipole). >>> >>> >>> Thanks a lot for any comment! >>> Best, >>> Max >>> >>> dipfit >>> >>> >>> >>> >>> >>> _______________________________________________ >>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html >>> To unsubscribe, send an empty email to >>> eeglablist-unsubscribe at sccn.ucsd.edu >>> For digest mode, send an email with the subject "set digest mime" to >>> eeglablist-request at sccn.ucsd.edu >>> >> >> >> >> -- >> Scott Makeig, Research Scientist and Director, Swartz Center for >> Computational Neuroscience, Institute for Neural Computation; Prof. of >> Neurosciences (Adj.), University of California San Diego, La Jolla CA >> 92093-0559, http://sccn.ucsd.edu/~scott >> >> _______________________________________________ >> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html >> To unsubscribe, send an empty email to >> eeglablist-unsubscribe at sccn.ucsd.edu >> For digest mode, send an email with the subject "set digest mime" to >> eeglablist-request at sccn.ucsd.edu >> >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Fri Aug 24 19:51:15 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Fri, 24 Aug 2012 19:51:15 +0200 Subject: [FieldTrip] ft_databrowser Message-ID: Hi, Does anybody knows why I can't see whole head topography in ft_databrowser cfg = []; cfg.viewmode ='component'; cfg.layout = 'CTF275.lay'; ft_databrowser(cfg, comp); [image: Inline images 1] Kind regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.jpeg Type: image/jpeg Size: 78272 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ica.JPG Type: image/jpeg Size: 78272 bytes Desc: not available URL: From r.vandermeij at donders.ru.nl Sat Aug 25 11:40:08 2012 From: r.vandermeij at donders.ru.nl (Roemer van der Meij) Date: Sat, 25 Aug 2012 11:40:08 +0200 Subject: [FieldTrip] ft_databrowser In-Reply-To: References: Message-ID: Hi Nenad, This was indeed a bug, and was fixed very recently. If you update to the latest version, it should work fine. Do note the speed-upgrade that I implemented ;), it should be much faster now. Also, just to gather some interest, when you now make a data-selection in the browser, you can right-click to get a context menu with several interesting options, do have a look! :) Best, Roemer On Fri, Aug 24, 2012 at 7:51 PM, Nenad Polomac wrote: > Hi, > > Does anybody knows why I can't see whole head topography in ft_databrowser > > > cfg = []; > cfg.viewmode ='component'; > cfg.layout = 'CTF275.lay'; > ft_databrowser(cfg, comp); > > [image: Inline images 1] > > Kind regards! > > Nenad > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- Roemer van der Meij M.Sc. PhD student Donders Institute for Brain, Cognition and Behaviour Centre for Cognition P.O. Box 9104 6500 HE Nijmegen The Netherlands Tel: +31(0)24 3655932 E-mail: r.vandermeij at donders.ru.nl -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.jpeg Type: image/jpeg Size: 78272 bytes Desc: not available URL: From polomacnenad at gmail.com Sat Aug 25 20:03:09 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Sat, 25 Aug 2012 20:03:09 +0200 Subject: [FieldTrip] fieldtrip Digest, Vol 21, Issue 31 In-Reply-To: References: Message-ID: Dear Roemer, Thank you very much for your suggestion. Now works fine and much faster. Keep up the good work! :) All the best! Nenad > Hi Nenad, > > This was indeed a bug, and was fixed very recently. If you update to the > latest version, it should work fine. Do note the speed-upgrade that I > implemented ;), it should be much faster now. > Also, just to gather some interest, when you now make a data-selection in > the browser, you can right-click to get a context menu with several > interesting options, do have a look! :) > > Best, > Roemer > > > > On Fri, Aug 24, 2012 at 7:51 PM, Nenad Polomac >wrote: > > > Hi, > > > > Does anybody knows why I can't see whole head topography in > ft_databrowser > > > > > > cfg = []; > > cfg.viewmode ='component'; > > cfg.layout = 'CTF275.lay'; > > ft_databrowser(cfg, comp); > > > > [image: Inline images 1] > > > > Kind regards! > > > > Nenad > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > > > -- > Roemer van der Meij M.Sc. > PhD student > Donders Institute for Brain, Cognition and Behaviour > Centre for Cognition > P.O. Box 9104 > 6500 HE Nijmegen > The Netherlands > Tel: +31(0)24 3655932 > E-mail: r.vandermeij at donders.ru.nl > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: < > http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20120825/351f7b56/attachment.html > > > -------------- next part -------------- > A non-text attachment was scrubbed... > Name: not available > Type: image/jpeg > Size: 78272 bytes > Desc: not available > URL: < > http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20120825/351f7b56/attachment.jpe > > > > ------------------------------ > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > End of fieldtrip Digest, Vol 21, Issue 31 > ***************************************** > -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Sat Aug 25 20:09:30 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Sat, 25 Aug 2012 20:09:30 +0200 Subject: [FieldTrip] ft_databrowser Message-ID: Dear Roemer, Thank you very much for your suggestion. Now works fine and much faster. Keep up the good work! :) I am sorry but I don't know how to put replay massage in the same thread in the Fieldtrip discussion list . Please give me some clue All the best! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From ayelet.landau at gmail.com Sat Aug 25 21:03:46 2012 From: ayelet.landau at gmail.com (Ayelet Landau) Date: Sat, 25 Aug 2012 21:03:46 +0200 Subject: [FieldTrip] ft_databrowser In-Reply-To: References: Message-ID: Hi all, I thought I would report here also another (perhaps related?) strange behavior that I encountered with databrowser. If you look at the attached image you will immediately note that there is an empty portion after the end of my data. The data plotted is indeed all the data (so, no missing data), but this extra white just appears to varying degrees from trial to trial. any idea why this occurs? thanks! Ayelet [image: Inline image 1] On Sat, Aug 25, 2012 at 8:09 PM, Nenad Polomac wrote: > > Dear Roemer, > > Thank you very much for your suggestion. Now works fine and much faster. > Keep up the good work! :) > > I am sorry but I don't know how to put replay massage in the same thread > in the Fieldtrip discussion list . Please give me some clue > > All the best! > > Nenad > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- Ayelet N. Landau Postdoctoral Scientist, Ernst Strüngmann Institute (ESI) for Neuroscience in Cooperation with Max Planck Society Deutschordenstr. 46, D-60528 Frankfurt Mobile: +49 (0)16 22733 110 Fax: +49 (0)69 96769 555 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.png Type: image/png Size: 44638 bytes Desc: not available URL: From r.vandermeij at donders.ru.nl Sun Aug 26 01:06:28 2012 From: r.vandermeij at donders.ru.nl (Roemer van der Meij) Date: Sun, 26 Aug 2012 01:06:28 +0200 Subject: [FieldTrip] ft_databrowser In-Reply-To: References: Message-ID: Hi Ayelet, This is an intended feature, to make sure each trial is plotted at the same 'zoom-level' (keeping the plotted size of the time-axis the same). We hope this will help in recognizing odd patterns and such. You can also now zoom-in into the time-axis by pressing shift+left or right, or by using the buttons at the lower left corner. Hope this info helps, Best, Roemer On Sat, Aug 25, 2012 at 9:03 PM, Ayelet Landau wrote: > Hi all, > I thought I would report here also another (perhaps related?) strange > behavior that I encountered with databrowser. If you look at the attached > image you will immediately note that there is an empty portion after the > end of my data. The data plotted is indeed all the data (so, no missing > data), but this extra white just appears to varying degrees from trial to > trial. > > any idea why this occurs? > > thanks! > Ayelet > > [image: Inline image 1] > > On Sat, Aug 25, 2012 at 8:09 PM, Nenad Polomac wrote: > >> >> Dear Roemer, >> >> Thank you very much for your suggestion. Now works fine and much faster. >> Keep up the good work! :) >> >> I am sorry but I don't know how to put replay massage in the same thread >> in the Fieldtrip discussion list . Please give me some clue >> >> All the best! >> >> Nenad >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip >> > > > > -- > Ayelet N. Landau > > Postdoctoral Scientist, > Ernst Strüngmann Institute (ESI) for Neuroscience > in Cooperation with Max Planck Society > Deutschordenstr. 46, D-60528 Frankfurt > > Mobile: +49 (0)16 22733 110 > Fax: +49 (0)69 96769 555 > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- Roemer van der Meij M.Sc. PhD student Donders Institute for Brain, Cognition and Behaviour Centre for Cognition P.O. Box 9104 6500 HE Nijmegen The Netherlands Tel: +31(0)24 3655932 E-mail: r.vandermeij at donders.ru.nl -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.png Type: image/png Size: 44638 bytes Desc: not available URL: From karl.doron at gmail.com Sun Aug 26 20:53:53 2012 From: karl.doron at gmail.com (Karl Doron) Date: Sun, 26 Aug 2012 11:53:53 -0700 Subject: [FieldTrip] connectivity/amplitude correlation Message-ID: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> Hello, I'm wondering if there are any publications with the maths behind the functions for ft_connectivity_corr. I would specifically like to confirm (for a pending publication) how the amplitude correlation is computed when the function is given output from ft_freqanalysis which contains a time dimension (relevant cfg parameters pasted below). I've assumed this is occurring in the frequency domain across trials for sensor i and sensor j. cfg.channelcmb = chancmb; cfg.method = 'mtmconvol'; cfg.output = 'fourier'; cfg.keeptrials = 'yes'; cfg.foi = band(1):band(2); cfg.toi = 1.8:0.01:3.0; Thanks for any feedback! Karl Doron, Ph.D. UC, Santa Barbara -------------- next part -------------- An HTML attachment was scrubbed... URL: From stan.vanpelt at fcdonders.ru.nl Mon Aug 27 09:59:02 2012 From: stan.vanpelt at fcdonders.ru.nl (Stan van Pelt) Date: Mon, 27 Aug 2012 09:59:02 +0200 (CEST) Subject: [FieldTrip] ft_databrowser In-Reply-To: Message-ID: <1604964460.83554.1346054342785.JavaMail.root@sculptor.zimbra.ru.nl> Hi Roemer, Interesting features! In that respect, it might also be useful to set the x-axis ticklabels at 'logical'/more intuitive values (e.g., 0.1s apart) instead of 1/xth of the total data length. Best, Stan ----- Oorspronkelijk bericht ----- > Van: "Roemer van der Meij" > Aan: "ayelet landau" , "FieldTrip discussion > list" > Verzonden: Zondag 26 augustus 2012 01:06:28 > Onderwerp: Re: [FieldTrip] ft_databrowser > Hi Ayelet, > This is an intended feature, to make sure each trial is plotted at the > same 'zoom-level' (keeping the plotted size of the time-axis the > same). We hope this will help in recognizing odd patterns and such. > You can also now zoom-in into the time-axis by pressing shift+left or > right, or by using the buttons at the lower left corner. > Hope this info helps, > Best, > Roemer > On Sat, Aug 25, 2012 at 9:03 PM, Ayelet Landau < > ayelet.landau at gmail.com > wrote: > > Hi all, > > I thought I would report here also another (perhaps related?) > > strange > > behavior that I encountered with databrowser. If you look at the > > attached image you will immediately note that there is an empty > > portion after the end of my data. The data plotted is indeed all the > > data (so, no missing data), but this extra white just appears to > > varying degrees from trial to trial. > > any idea why this occurs? > > thanks! > > Ayelet > > Inline image 1 > > On Sat, Aug 25, 2012 at 8:09 PM, Nenad Polomac < > > polomacnenad at gmail.com > wrote: > > > Dear Roemer, > > > Thank you very much for your suggestion. Now works fine and much > > > faster. > > > Keep up the good work! :) > > > I am sorry but I don't know how to put replay massage in the same > > > thread in the Fieldtrip discussion list . Please give me some clue > > > All the best! > > > Nenad > > > _______________________________________________ > > > fieldtrip mailing list > > > fieldtrip at donders.ru.nl > > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > -- > > Ayelet N. Landau > > Postdoctoral Scientist, > > Ernst Strüngmann Institute (ESI) for Neuroscience > > in Cooperation with Max Planck Society > > Deutschordenstr. 46, D-60528 Frankfurt > > Mobile: +49 (0)16 22733 110 > > Fax: +49 (0)69 96769 555 > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -- > Roemer van der Meij M.Sc. > PhD student > Donders Institute for Brain, Cognition and Behaviour > Centre for Cognition > P.O. Box 9104 > 6500 HE Nijmegen > The Netherlands > Tel: +31(0)24 3655932 > E-mail: r.vandermeij at donders.ru.nl > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Stan van Pelt Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen Kapittelweg 29, 6525 EN Nijmegen, Netherlands E-mail: stan.vanpelt at donders.ru.nl Website: www.ru.nl/donders/ Phone: (+31) (0)24 36 10981 Fax: (+31) (0)24 36 10989 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.png Type: image/png Size: 44638 bytes Desc: not available URL: From Lilla.Magyari at mpi.nl Mon Aug 27 11:59:53 2012 From: Lilla.Magyari at mpi.nl (Magyari, Lilla) Date: Mon, 27 Aug 2012 11:59:53 +0200 Subject: [FieldTrip] connectivity/amplitude correlation In-Reply-To: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> References: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> Message-ID: <503B4519.5030506@mpi.nl> Dear Karl, There are four papers mentioned in the help of ft_connectivity_corr. I am wondering if any of those could answer your question. Best, Lilla Karl Doron wrote: > Hello, > > I'm wondering if there are any publications with the maths behind the > functions for ft_connectivity_corr. I would specifically like to confirm > (for a pending publication) how the amplitude correlation is computed > when the function is given output from ft_freqanalysis which contains a > time dimension (relevant cfg parameters pasted below). I've assumed this > is occurring in the frequency domain across trials for sensor i and > sensor j. > > cfg.channelcmb = chancmb; > cfg.method = 'mtmconvol'; > cfg.output = 'fourier'; > cfg.keeptrials = 'yes'; > cfg.foi = band(1):band(2); > cfg.toi = 1.8:0.01:3.0; > > > Thanks for any feedback! > > > Karl Doron, Ph.D. > UC, Santa Barbara > > > > ------------------------------------------------------------------------ > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From karl.doron at gmail.com Mon Aug 27 16:16:52 2012 From: karl.doron at gmail.com (Karl Doron) Date: Mon, 27 Aug 2012 07:16:52 -0700 Subject: [FieldTrip] connectivity/amplitude correlation In-Reply-To: <503B4519.5030506@mpi.nl> References: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> <503B4519.5030506@mpi.nl> Message-ID: <0F563C8B-4EA4-43F2-9FF3-93625C9E4A1D@gmail.com> Hi Lilla, There is a reference for Rosenberg et al, 1998 for the partialization method in the code itself but I'm not seeing any in the help section. Thanks, karl On Aug 27, 2012, at 2:59 AM, "Magyari, Lilla" wrote: > Dear Karl, > > There are four papers mentioned in the help of ft_connectivity_corr. I am wondering if any of those could answer your question. > > Best, > Lilla > > Karl Doron wrote: >> Hello, >> I'm wondering if there are any publications with the maths behind the functions for ft_connectivity_corr. I would specifically like to confirm (for a pending publication) how the amplitude correlation is computed when the function is given output from ft_freqanalysis which contains a time dimension (relevant cfg parameters pasted below). I've assumed this is occurring in the frequency domain across trials for sensor i and sensor j. >> cfg.channelcmb = chancmb; >> cfg.method = 'mtmconvol'; >> cfg.output = 'fourier'; >> cfg.keeptrials = 'yes'; >> cfg.foi = band(1):band(2); >> cfg.toi = 1.8:0.01:3.0; >> Thanks for any feedback! >> Karl Doron, Ph.D. >> UC, Santa Barbara >> ------------------------------------------------------------------------ >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From Lilla.Magyari at mpi.nl Mon Aug 27 16:56:37 2012 From: Lilla.Magyari at mpi.nl (Magyari, Lilla) Date: Mon, 27 Aug 2012 16:56:37 +0200 Subject: [FieldTrip] connectivity/amplitude correlation In-Reply-To: <0F563C8B-4EA4-43F2-9FF3-93625C9E4A1D@gmail.com> References: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> <503B4519.5030506@mpi.nl> <0F563C8B-4EA4-43F2-9FF3-93625C9E4A1D@gmail.com> Message-ID: <503B8AA5.6020207@mpi.nl> hi Karl, Maybe your version of FieldTrip is older, and therefore, it does not contain the references yet. You can check the reference documentation of ft_connectivity_corr on the FieldTrip wiki: http://fieldtrip.fcdonders.nl/reference/ft_connectivity_corr (The reference documentation of the functions is always automatically updated according to the latest version of the documentation part (or "help" part) of the script.) The same articles (and more) can be found also under Literature (Method References) http://fieldtrip.fcdonders.nl/references_to_implemented_methods under the title Connectivity Analysis. I have not read the articles myself, so I am not sure if they give the information you need. But I was just wondering if you are aware of the recently available documentation of the function. Best, Lilla Karl Doron wrote: > Hi Lilla, > > There is a reference for Rosenberg et al, 1998 for the partialization method in the code itself but I'm not seeing any in the help section. > > Thanks, > karl > > > On Aug 27, 2012, at 2:59 AM, "Magyari, Lilla" wrote: > >> Dear Karl, >> >> There are four papers mentioned in the help of ft_connectivity_corr. I am wondering if any of those could answer your question. >> >> Best, >> Lilla >> >> Karl Doron wrote: >>> Hello, >>> I'm wondering if there are any publications with the maths behind the functions for ft_connectivity_corr. I would specifically like to confirm (for a pending publication) how the amplitude correlation is computed when the function is given output from ft_freqanalysis which contains a time dimension (relevant cfg parameters pasted below). I've assumed this is occurring in the frequency domain across trials for sensor i and sensor j. >>> cfg.channelcmb = chancmb; >>> cfg.method = 'mtmconvol'; >>> cfg.output = 'fourier'; >>> cfg.keeptrials = 'yes'; >>> cfg.foi = band(1):band(2); >>> cfg.toi = 1.8:0.01:3.0; >>> Thanks for any feedback! >>> Karl Doron, Ph.D. >>> UC, Santa Barbara >>> ------------------------------------------------------------------------ >>> _______________________________________________ >>> fieldtrip mailing list >>> fieldtrip at donders.ru.nl >>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From r.oostenveld at donders.ru.nl Tue Aug 28 10:05:25 2012 From: r.oostenveld at donders.ru.nl (Robert Oostenveld) Date: Tue, 28 Aug 2012 10:05:25 +0200 Subject: [FieldTrip] PLV on sources In-Reply-To: <201208241235.q7OCZhaZ002930@brievenbus.science.ru.nl> References: <201208241235.q7OCZhaZ002930@brievenbus.science.ru.nl> Message-ID: Dear Robert and Nayaran, The data handling in ft_connectivityanalysis is not yet at the level that it can easily deal with all source representations; that is something that we will improve upon in the future. The way we suggest it now is similar to your suggestion and is outlined in detail in the http://fieldtrip.fcdonders.nl/tutorial/connectivity tutorial. Specifically have a look at the second section where as example source (virtual channel) coherence and granger is computed. In that tutorial you would just change cfg.method in plv and you should be rolling. Best regards, Robert O. PS this connectivity tutorial is quite new, so might also be of interest for other people doing connectivity analysis! Op 24 aug. 2012 om 14:35 heeft "Narayan Puthanmadam Subramaniyam" het volgende geschreven: > hi > > i was thinking of doing the same thing exactly. Computing plv on the eeg inverse solution. technically it should be possible going by the general level concept of phase locking. but i am not sure so let the experts here speak. > > Thanks&Regards, > NPS > > Sent from my Nokia phone > -----Original Message----- > From: Robert Brown > Sent: 24:08:2012, 11:38 > To: fieldtrip at science.ru.nl > Subject: [FieldTrip] PLV on sources > > > Dear Fieldtrippers, > > I see (from the ft_connectivityanalysis code) that one cannot run phase > locking (PLV) analysis (a la Lachaux et al) on source data. > > My brain still insisted on trying to do so but unfortunately I am not sure > if it is clever enough. Hope you can help! > > It seems that a way to go would be to do LCMV beamforming on the data, get > the virtual electrodes, then run freqanalysis with fourier output on these > virtual electrodes and then perform PLV on those data. > > Before I dig myself very deep into this, maybe some experts here could > guide me: Would this approach work? Any possible caveats? Are there maybe > other, better and more straightforward ways for achieving the PLV on source > level? > > Any comments would be appreciated. Thank you very much for your time! > > Kind regards, > Bob > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip From sherrykhan78 at gmail.com Tue Aug 28 12:34:14 2012 From: sherrykhan78 at gmail.com (Sheraz Khan) Date: Tue, 28 Aug 2012 06:34:14 -0400 Subject: [FieldTrip] PLV on sources In-Reply-To: References: <201208241235.q7OCZhaZ002930@brievenbus.science.ru.nl> Message-ID: Dear Bob, Another possible solution is to do wavelet decomposition in sensor space and then multiply it with your imaging kernel, then do PLV in source space, this will save some computational time, but still you have to deal with EEG/MEG point spread function. Sheraz Khan Martinos Center MGH/MIT/Harvard On Tue, Aug 28, 2012 at 4:05 AM, Robert Oostenveld < r.oostenveld at donders.ru.nl> wrote: > Dear Robert and Nayaran, > > The data handling in ft_connectivityanalysis is not yet at the level that > it can easily deal with all source representations; that is something that > we will improve upon in the future. > > The way we suggest it now is similar to your suggestion and is outlined in > detail in the http://fieldtrip.fcdonders.nl/tutorial/connectivitytutorial. Specifically have a look at the second section where as example > source (virtual channel) coherence and granger is computed. In that > tutorial you would just change cfg.method in plv and you should be rolling. > > Best regards, > Robert O. > > PS this connectivity tutorial is quite new, so might also be of interest > for other people doing connectivity analysis! > > > Op 24 aug. 2012 om 14:35 heeft "Narayan Puthanmadam Subramaniyam" < > narayan.ps at tut.fi> het volgende geschreven: > > > hi > > > > i was thinking of doing the same thing exactly. Computing plv on the eeg > inverse solution. technically it should be possible going by the general > level concept of phase locking. but i am not sure so let the experts here > speak. > > > > Thanks&Regards, > > NPS > > > > Sent from my Nokia phone > > -----Original Message----- > > From: Robert Brown > > Sent: 24:08:2012, 11:38 > > To: fieldtrip at science.ru.nl > > Subject: [FieldTrip] PLV on sources > > > > > > Dear Fieldtrippers, > > > > I see (from the ft_connectivityanalysis code) that one cannot run phase > > locking (PLV) analysis (a la Lachaux et al) on source data. > > > > My brain still insisted on trying to do so but unfortunately I am not > sure > > if it is clever enough. Hope you can help! > > > > It seems that a way to go would be to do LCMV beamforming on the data, > get > > the virtual electrodes, then run freqanalysis with fourier output on > these > > virtual electrodes and then perform PLV on those data. > > > > Before I dig myself very deep into this, maybe some experts here could > > guide me: Would this approach work? Any possible caveats? Are there maybe > > other, better and more straightforward ways for achieving the PLV on > source > > level? > > > > Any comments would be appreciated. Thank you very much for your time! > > > > Kind regards, > > Bob > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > -------------- next part -------------- An HTML attachment was scrubbed... URL: From r.oostenveld at donders.ru.nl Thu Aug 30 10:31:17 2012 From: r.oostenveld at donders.ru.nl (Robert Oostenveld) Date: Thu, 30 Aug 2012 10:31:17 +0200 Subject: [FieldTrip] PLV on sources In-Reply-To: References: <201208241235.q7OCZhaZ002930@brievenbus.science.ru.nl> Message-ID: Dear Sheraz, Bob You are completely right; since the fft and the spatial filter multiplication are both linear operations, their order can be interchanged. If nsource>nchan, this will speed things up. For full brain source reconstructions it will be much faster (and require less memory). It would be nice to extend the tutorial with a section that explains precisely this. Best Robert Op 28 aug. 2012 om 12:34 heeft Sheraz Khan het volgende geschreven: > Dear Bob, > > Another possible solution is to do wavelet decomposition in sensor space and then multiply it with your imaging kernel, then do PLV in source space, this will save some computational time, but still you have to deal with EEG/MEG point spread function. > > Sheraz Khan > Martinos Center > MGH/MIT/Harvard > > On Tue, Aug 28, 2012 at 4:05 AM, Robert Oostenveld wrote: > Dear Robert and Nayaran, > > The data handling in ft_connectivityanalysis is not yet at the level that it can easily deal with all source representations; that is something that we will improve upon in the future. > > The way we suggest it now is similar to your suggestion and is outlined in detail in the http://fieldtrip.fcdonders.nl/tutorial/connectivity tutorial. Specifically have a look at the second section where as example source (virtual channel) coherence and granger is computed. In that tutorial you would just change cfg.method in plv and you should be rolling. > > Best regards, > Robert O. > > PS this connectivity tutorial is quite new, so might also be of interest for other people doing connectivity analysis! > > > Op 24 aug. 2012 om 14:35 heeft "Narayan Puthanmadam Subramaniyam" het volgende geschreven: > > > hi > > > > i was thinking of doing the same thing exactly. Computing plv on the eeg inverse solution. technically it should be possible going by the general level concept of phase locking. but i am not sure so let the experts here speak. > > > > Thanks&Regards, > > NPS > > > > Sent from my Nokia phone > > -----Original Message----- > > From: Robert Brown > > Sent: 24:08:2012, 11:38 > > To: fieldtrip at science.ru.nl > > Subject: [FieldTrip] PLV on sources > > > > > > Dear Fieldtrippers, > > > > I see (from the ft_connectivityanalysis code) that one cannot run phase > > locking (PLV) analysis (a la Lachaux et al) on source data. > > > > My brain still insisted on trying to do so but unfortunately I am not sure > > if it is clever enough. Hope you can help! > > > > It seems that a way to go would be to do LCMV beamforming on the data, get > > the virtual electrodes, then run freqanalysis with fourier output on these > > virtual electrodes and then perform PLV on those data. > > > > Before I dig myself very deep into this, maybe some experts here could > > guide me: Would this approach work? Any possible caveats? Are there maybe > > other, better and more straightforward ways for achieving the PLV on source > > level? > > > > Any comments would be appreciated. Thank you very much for your time! > > > > Kind regards, > > Bob > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > > _______________________________________________ > > fieldtrip mailing list > > fieldtrip at donders.ru.nl > > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -------------- next part -------------- An HTML attachment was scrubbed... URL: From polomacnenad at gmail.com Thu Aug 30 10:33:28 2012 From: polomacnenad at gmail.com (Nenad Polomac) Date: Thu, 30 Aug 2012 10:33:28 +0200 Subject: [FieldTrip] neighbours Message-ID: Hi, I am analyzing data obtained on CTF MEG 275 system. However, 4 sensors are broken and data ended up with 271 sensor. So, my question is should I interpolate the missing sensors with FT_CHANNELREPAI *R*? In the case I leave data with 271 sensor, will those missing sensors influence neighbors configuration in the statistic and the source localisation analysis? Thank you in advance! Kind regards! Nenad -------------- next part -------------- An HTML attachment was scrubbed... URL: From jan.schoffelen at donders.ru.nl Thu Aug 30 21:57:16 2012 From: jan.schoffelen at donders.ru.nl (jan-mathijs schoffelen) Date: Thu, 30 Aug 2012 21:57:16 +0200 Subject: [FieldTrip] connectivity/amplitude correlation In-Reply-To: <0F563C8B-4EA4-43F2-9FF3-93625C9E4A1D@gmail.com> References: <6A2AC74A-8085-4207-BA66-72A87A4C5B0C@gmail.com> <503B4519.5030506@mpi.nl> <0F563C8B-4EA4-43F2-9FF3-93625C9E4A1D@gmail.com> Message-ID: <274B2F2C-FE26-4A89-A2EC-84405BB77C2F@donders.ru.nl> Dear Karl, Perhaps I can add one or two things to Lilla's reply. If I understand your question correctly, it pertains specifically to the math behind the amplitude correlation computation, and how this is done in ft_connectivity_corr. It is indeed as you thought, the dimension over which the correlation is computed is indeed the trial dimension. That is, for each channelpair-time-frequency-point the amplitude correlation is computed across trials. A lot of the intelligence already takes place in ft_connectivityanalysis. In your case ft_connectivityanalysis computes the covariance (and variance) across trials between the amplitudes. The only thing that happens in ft_connectivity_corr is the normalisation with the square root of the variance product. Hope this helps, Jan-Mathijs On Aug 27, 2012, at 4:16 PM, Karl Doron wrote: > Hi Lilla, > > There is a reference for Rosenberg et al, 1998 for the partialization method in the code itself but I'm not seeing any in the help section. > > Thanks, > karl > > > On Aug 27, 2012, at 2:59 AM, "Magyari, Lilla" wrote: > >> Dear Karl, >> >> There are four papers mentioned in the help of ft_connectivity_corr. I am wondering if any of those could answer your question. >> >> Best, >> Lilla >> >> Karl Doron wrote: >>> Hello, >>> I'm wondering if there are any publications with the maths behind the functions for ft_connectivity_corr. I would specifically like to confirm (for a pending publication) how the amplitude correlation is computed when the function is given output from ft_freqanalysis which contains a time dimension (relevant cfg parameters pasted below). I've assumed this is occurring in the frequency domain across trials for sensor i and sensor j. >>> cfg.channelcmb = chancmb; >>> cfg.method = 'mtmconvol'; >>> cfg.output = 'fourier'; >>> cfg.keeptrials = 'yes'; >>> cfg.foi = band(1):band(2); >>> cfg.toi = 1.8:0.01:3.0; >>> Thanks for any feedback! >>> Karl Doron, Ph.D. >>> UC, Santa Barbara >>> ------------------------------------------------------------------------ >>> _______________________________________________ >>> fieldtrip mailing list >>> fieldtrip at donders.ru.nl >>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip >> _______________________________________________ >> fieldtrip mailing list >> fieldtrip at donders.ru.nl >> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip Jan-Mathijs Schoffelen, MD PhD Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University Nijmegen, The Netherlands Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands J.Schoffelen at donders.ru.nl Telephone: +31-24-3614793 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jm.horschig at donders.ru.nl Fri Aug 31 09:50:17 2012 From: jm.horschig at donders.ru.nl (=?ISO-8859-1?Q?=22J=F6rn_M=2E_Horschig=22?=) Date: Fri, 31 Aug 2012 09:50:17 +0200 Subject: [FieldTrip] neighbours In-Reply-To: References: Message-ID: <50406CB9.7090209@donders.ru.nl> Hi Nenad, the question you are asking does not deserve a definite 'yes' or 'no'. In any case, note that when averaging over subjects, missing sensors for one subject will be removed for all other subjects as well. So, if you are talking about different missing sensors per subject, an interpolation will definitely be a wise thing to do. In case you choose for interpolating the missing channels, you can try to check out the new spherical spline interpolation method (see help ft_channelrepair) - theoretically that should result in a quite good reconstruction. But the (standard) nearest neighbour interpolation is fine as well. With respect to your second question: missing sensors will influence the neighbour configuration depending on the method you chose for neighbour selection. For the 'distance' and 'template' method, each neighbouring channel of the missing will have one sensor less (i.e. the missing one). The 'triangulation' method will create triangle neighbour no matter how many channels you have, so it will just ignore the whole in space. In any case, I'd propose that you check out how satisfied you personally are with the neighbours by calling ft_neighbourplot. Just last week, I added the new option that if you use cfg.enableedit='yes', you can interactively change the neighbourhood structure within the neighbourplot. If I were you, I would start from the template method. As an additional piece of information: I am currently working on improving the neighbour templates, and will upload them probably next week. All that advertised, let me also say that having 4 out of 275 sensors missing won't affect your statistics much. Here, it of course depends on your region if interest, research question etc. E.g. if you are interested in posterior alpha power, but miss four temporal channels, there is no reason to worry. So, without knowing what you are looking for and where the missing channels are, I cannot give you a general advise what to do. As a last remark, the sensitiviy of the statistics will of course be influenced by missing sensors, but I doubt that it will matter much. Btw, I am not quite sure how much the spherical spline interpolation will buy you here, you might give it a try. I bet that it will increase your sensitivity more than the nearest neighbour approach, although both might result in rather low increases. But this is just a general gut feeling rather than anything I empirically validated. Best, Jörn On 8/30/2012 10:33 AM, Nenad Polomac wrote: > Hi, > > I am analyzing data obtained on CTF MEG 275 system. However, 4 sensors > are broken and data ended up with 271 sensor. So, my question is > should I interpolate the missing sensors with FT_CHANNELREPAI > _R_? In the > case I leave data with 271 sensor, will those missing sensors > influence neighbors configuration in the statistic and the source > localisation analysis? > > Thank you in advance! > > Kind regards! > > Nenad > > > _______________________________________________ > fieldtrip mailing list > fieldtrip at donders.ru.nl > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip -- Jörn M. Horschig PhD Student Donders Institute for Brain, Cognition and Behaviour Centre for Cognitive Neuroimaging Radboud University Nijmegen Neuronal Oscillations Group FieldTrip Development Team P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Contact: E-Mail: jm.horschig at donders.ru.nl Tel: +31-(0)24-36-68493 Web: http://www.ru.nl/donders Visiting address: Trigon, room 2.30 Kapittelweg 29 NL-6525 EN Nijmegen The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From akiko.ikkai at gmail.com Fri Aug 31 21:59:00 2012 From: akiko.ikkai at gmail.com (Akiko Ikkai) Date: Fri, 31 Aug 2012 15:59:00 -0400 Subject: [FieldTrip] FT_SCALPCURRENTDENSITY options and elec units Message-ID: Hi, I'm trying to run plv analysis, first acquiring SCD of my data with ft_scalcurrentdensity, and am confused about the difference between cfg.method = 'finite' and 'spline.' Both gave similar results short-range, but 'spline' gives higher plv at long range plv. I'd love it if someone could explain or suggest reading on the difference between these two methods (or is one method preferred over the other in some cases?). Also, I'm not sure whether or not I should convert my elec.pnt values, since the code ft_scalcurrentdensity says that "Note that the skin conductivity, electrode dimensions and the potential all have to be expressed in the same SI units." My elec.pnt are in cm, and I'm using the default conductivity (0.33 S/m). Do I need to do something like elec.pnt = elec.pnt/100 to convert the unit from cm to m? With or without conversion, I got the same results (viewed with ft_topoplotTFR), and I'm a little confused... Thank you in advance for your time! Akiko -- Akiko Ikkai, Ph.D. Postdoctoral Fellow Department of Psychological and Brain Sciences Johns Hopkins University Ames Hall, 3400 N. Charles St. Baltimore, MD 21218 -------------- next part -------------- An HTML attachment was scrubbed... URL: