<div dir="ltr"><div><div><div><div>Thanks you <br><br></div>Your code works fine. Here is the Matlab output for this code:<br><i><br>the input is timelock data with 2 channels and 5 timebins<br>removing confound 1 2 3 <br>
keeping confound <br>normalizing the confounds, except the constant <br>estimating the regression weights and removing the confounds <br>updating descriptives <br>the input is timelock data with 2 channels and 5 timebins<br>
Warning: the data does not contain a trial definition <br>> In utilities\private\warning_once at 158<br> In utilities\private\fixsampleinfo at 66<br> In ft_datatype_raw at 154<br> In ft_checkdata at 298<br> In ft_timelockanalysis at 105<br>
In ft_regressconfound at 313<br>Warning: reconstructing sampleinfo by assuming that the trials are consecutive segments of a continuous<br>recording <br>> In utilities\private\warning_once at 158<br> In utilities\private\fixsampleinfo at 79<br>
In ft_datatype_raw at 154<br> In ft_checkdata at 298<br> In ft_timelockanalysis at 105<br> In ft_regressconfound at 313<br>averaging trials<br>averaging trial 20 of 20<br><br>the call to "ft_timelockanalysis" took 1 seconds<br>
the call to "ft_regressconfound" took 4 seconds</i><br><br></div>The data I am trying to correct for the motion is real. It was collected with a current phantom generating the signal by a dipole. The data is continuous but can be broken into trials of 1 sec (the dipole generates 7 Hz sinusoidal signal continuously). The phantom was mechanically moved during the scan for a precisely known amount (e.g. 1 cm down). The data was recorded by CTF 275-sensor MEG unit with the continuous head coils localization. I was able to apply the SSS motion correction algorithm successfully to this data set. If there is a way I could share this data set.<br>
<br></div>Thanks,<br></div>Inna<br></div><div class="gmail_extra"><br clear="all"><div>Inna McGowin<br></div>
<br><br><div class="gmail_quote">On Mon, Dec 9, 2013 at 9:48 AM, Arjen Stolk <span dir="ltr"><<a href="mailto:a.stolk8@gmail.com" target="_blank">a.stolk8@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div dir="ltr"><div><div><div>Hi Inna,<br><br></div>Testing ft_regressconfound with some fake data, everything seems to be fine here:<br><br>timelock = [];<br>timelock.label = {'1' '2'};<br>timelock.time = 1:5;<br>
timelock.dimord = 'rpt_chan_time';<br>timelock.trial = randn(20,2,5);<br>timelock.avg = randn(2,5);<br><br>cfg = [];<br>cfg.confound = randn(20,3);<br>cfg.reject = [1:3];<br>timelock_out = ft_regressconfound(cfg, timelock);<br>
<br></div>Does this example code work for you? I could not detect any mistake in your lines of code. What does your 'timelock' structure look like? Are there any real values in the data? (don't think this is causing it, but please update to newest Ft version just to be sure)<br>
<br></div>Yours,<br>Arjen<br><div><div><div><div><div><div><div><div class="gmail_extra"><br><br><div class="gmail_quote">2013/12/2 McGowin, Inna <span dir="ltr"><<a href="mailto:mcgoiv0@wfu.edu" target="_blank">mcgoiv0@wfu.edu</a>></span><div>
<div class="h5"><br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div><div><div>Thank you Arjen,<br></div><div>Actually, I am testing the FieldTrip motion regression on MEG data specifically collected with a current phantom and known motion. I can treat the data as ERP data and average over 1 sec trials if needed. <br>
</div>Following your advise on reduction of the regressors number (on 120 trials of 1 sec long) I was able to avoid the "out of memory" issue. Though I run into the different set of problems now that I need help to understand:<br>
______________________________________________<br>??? Error using ==> mxSerialize<br>Error during serialization of (null)<br><br>Error in ==> ft_postamble_provenance at 91<br> cfg.callinfo.outputhash{iargout} = CalcMD5(mxSerialize(tmparg));<br>
<br>Error in ==> ft_postamble at 55<br> evalin('caller', ['ft_postamble_' cmd]);<br><br>Error in ==> ft_timelockanalysis at 370<br>ft_postamble provenance timelock<br><br>Error in ==> ft_regressconfound at 313<br>
dataout = ft_timelockanalysis(tempcfg, dataout);% reaveraging<br>_________________________________________________________<br><br></div>Here is the code I run to regress the head motion:<br>%%<br>addpath C:\Userdata\MATLAB\FieldTrip\fieldtrip-20131023<br>
ft_defaults<br>cfg.dataset = 'InnaTest_Current-Phantom_20131004_09.ds';<br>cfg.trialdef.triallength=1; <br>cfg.trialdef.ntrials=120; <br>cfg.continuous = 'yes';<br>cfg = ft_definetrial(cfg); <br>cfg.channel = {'MEG'};<br>
cfg.demean = 'yes';<br>data = ft_preprocessing(cfg); <br><br>%%<br>cfg = [];<br>cfg.keeptrials = 'yes';<br>timelock = ft_timelockanalysis(cfg, data); <br><br>%%<br>cfg = [];<br>
cfg.dataset = 'InnaTest_Current-Phantom_20131004_09.ds';<br>cfg.trialdef.triallength=1;<br>cfg.trialdef.ntrials=120; <br>cfg.continuous = 'yes';<br>cfg = ft_definetrial(cfg);<br><br>%%<br>cfg.channel = {'HLC0011','HLC0012','HLC0013', ...<br>
'HLC0021','HLC0022','HLC0023', ...<br> 'HLC0031','HLC0032','HLC0033'};<br>headpos = ft_preprocessing(cfg);<br><br>
ntrials = length(headpos.sampleinfo)<br>for t = 1:ntrials<br>coil1(:,t) = [mean(headpos.trial{1,t}(1,:)); mean(headpos.trial{1,t}(2,:)); mean(headpos.trial{1,t}(3,:))];<br>coil2(:,t) = [mean(headpos.trial{1,t}(4,:)); mean(headpos.trial{1,t}(5,:)); mean(headpos.trial{1,t}(6,:))];<br>
coil3(:,t) = [mean(headpos.trial{1,t}(7,:)); mean(headpos.trial{1,t}(8,:)); mean(headpos.trial{1,t}(9,:))];<br>end<br><br>cc = circumcenter(coil1, coil2, coil3);<br>cc_dem = [cc - repmat(mean(cc,2),1,size(cc,2))]';<br>
<br>%%<br>confound = [cc_dem ...<br>ones(size(cc_dem,1),1)]; <br><br>%%<br>cfg = [];<br>cfg.confound = confound;<br>cfg.reject = [1:6];<br>regr = ft_regressconfound(cfg, timelock);<br>
%%<br><br>_________________________________________________________________<br></div>Sorry for the lengthy letter and thanks for the help!<span><font color="#888888"><br></font></span><div><div class="gmail_extra">
<span><font color="#888888"><br clear="all"><div>Inna<br></div><br><br><br><br>
<br></font></span><div class="gmail_quote"><div><div>On Thu, Nov 28, 2013 at 12:23 PM, Stolk, A. (Arjen) <span dir="ltr"><<a href="mailto:a.stolk@fcdonders.ru.nl" target="_blank">a.stolk@fcdonders.ru.nl</a>></span> wrote:<br>
</div></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div><div>
<div>Hi Inna,<br><br>Are you using the same design matrix as on the wiki page? That one covers the primary head translations and rotations, but also their derivatives. You could try using the primary measures only (3 translastions, and 3 rotations of the circumcenter, totaling 6 + 1 constant regressors). It is also mentionworthy that there should be a balance between the number of regressors and the number of observations (i.e. trials). I believe on the respective wiki page on head movement regression it states that the ratio is recommended to be not larger than 1:10 (regressors:trials), accompanied with a reference. Accordingly, I'd suggest to use those 7 regressors, and your data (120s resting state) split in segments of 1 or 2 seconds, totaling 60 or 120 trials. These specifications might hopefully also allow working around your memory issue.<br>
<br>Second, it may be worth mentioning what you would like to do with the data post-movement-regression. Namely, the movement compensation will remove contributions from head movement to the signal. But this requires that signal to behave consistently over trials, which does not seem the case with the signal in your resting study? That is, with an ERP study, the signal fluctuations appear consistent over trials, with amplitudes being modulated by head movement (i.e. sensor-source distance, see our neuroimage paper - 2013) in a predictable fashion. So if you're planning on doing frequency analysis, or maybe connectivity analysis, after head movement compensation, I'd recommend reversing that order. The rule of thumb is to use ft_regressconfound just prior to ft_xxxstatistics, removing otherwise unexplained variance (over trials) due to head movement, benefitting your statistical assessments. <br>
<br>Hope these suggestions may help you with further analyses. I'm not familiar with anyone implementing this method to resting state analyses, but maybe someone else has some first-hand experience here. Furthermore, too late for your current dataset, but for a next one I'd recommend using the online head position monitoring/respositioning tool that we have developed for CTF systems (a neuromag version is approaching the final stage). This tool is routinely used in our lab to monitor and reduce head movment throughout recording (after which ft_regressconfound is being used to deal with the trial-by-trial crumbles). <br>
<br>Yours,<br>Arjen</div><br></div></div><div>_______________________________________________<br>
fieldtrip mailing list<br>
<a href="mailto:fieldtrip@donders.ru.nl" target="_blank">fieldtrip@donders.ru.nl</a><br>
<a href="http://mailman.science.ru.nl/mailman/listinfo/fieldtrip" target="_blank">http://mailman.science.ru.nl/mailman/listinfo/fieldtrip</a><br></div></blockquote></div><br></div></div></div>
<br>_______________________________________________<br>
fieldtrip mailing list<br>
<a href="mailto:fieldtrip@donders.ru.nl" target="_blank">fieldtrip@donders.ru.nl</a><br>
<a href="http://mailman.science.ru.nl/mailman/listinfo/fieldtrip" target="_blank">http://mailman.science.ru.nl/mailman/listinfo/fieldtrip</a><br></blockquote></div></div></div><br></div></div></div></div></div></div></div>
</div></div>
<br>_______________________________________________<br>
fieldtrip mailing list<br>
<a href="mailto:fieldtrip@donders.ru.nl">fieldtrip@donders.ru.nl</a><br>
<a href="http://mailman.science.ru.nl/mailman/listinfo/fieldtrip" target="_blank">http://mailman.science.ru.nl/mailman/listinfo/fieldtrip</a><br></blockquote></div><br></div>