[FieldTrip] problem aligning MRI with gradiometer

Luisa Frei l.frei at psy.gla.ac.uk
Sat Feb 12 00:05:15 CET 2011


I'm not an expert on beamforming myself. But how clear your results  
are depends on the preprocessing as well. If your sources are very  
low in the head, you could have muscle or heartbeat artifacts. I did  
an ICA to remove the heartbeat component.
Sorry I can't give you any more help, someone else can probably help  
you more.

Luisa

On 11 Feb 2011, at 22:25, Allison Connolly wrote:

> Thanks that seemed to help the scaling issue. However, my  
> beamforming is still not producing results. There is a huge bias  
> towards the center of the head, even though I accounted for the  
> neural activity index. It seems like the beamforming is just  
> projecting everything inwards. I'm not sure where to go from here.  
> Do you have any suggestions on different configuration settings I  
> could try for the beamforming?
>
> -Allison
>
> On Fri, Feb 11, 2011 at 4:02 PM, Luisa Frei <l.frei at psy.gla.ac.uk>  
> wrote:
> Hi Alison,
> how do you change the units? I convert them using:
>
> data.grad=ft_convert_units(data.grad,'cm')
>
> Maybe this helps.
>
> Luisa
>
>
> On 11 Feb 2011, at 21:19, Allison Connolly wrote:
>
> Hello,
>
> I am trying to perform beamforming on some MEG data (148 channels).  
> I have the dicom files for the MRI, and I have converted them  
> to .img and .hdr files using Curry.
> I am importing them into fieldtrip via
> mri=ft_read_mri('MRI.img').
>
> Then I manually declaring the fiducials with
> cfg.method                  = 'interactive';
> mrialign                    = ft_volumerealign(cfg, mri);
>
> Then segmenting with
> cfg = [];
> cfg.write        = 'no';
> [segment] = ft_volumesegment(cfg, mri);
>
> Then creating the head model with
> cfg = [];
> cfg.smooth                  = 'no';
> vol = ft_prepare_singleshell(cfg, segment);
>
> Then creating the grid with
> cfg                 = [];
> cfg.grad            = freq.grad;
> cfg.vol             = vol;
> cfg.reducerank      = 'no';
> cfg.channel                 = {'MEG','-TRIGGER','-RESPONSE',...
>     '-UACurrent','-MxA','-MyA','-MzA','-GxxA','-GyxA','-GyyA',...
>     '-GzyA','-GzxA','-MxaA','-MyaA','-MzaA','-E1','-E2','-E3','-E4'};
> cfg.grid.xgrid              = [-10:0.5:15]; % adjust according to vol
> cfg.grid.ygrid              = [-15:0.5:15]; % adjust according to vol
> cfg.grid.zgrid              = [-10:0.5:15]; % adjust according to vol
> [grid] = ft_prepare_leadfield(cfg);
>
> I am importing my MEG data with
> cfg = [];  % empty configuration
> cfg.dataset                 = 'C:/MEG Study/Subjectd001/d001/ 
> Billy_180s/06 at -09@-10 at _19_41/4/c,rfhp0.1Hz';
> cfg.trialfun                = 'trialfun_general';
> cfg.trialdef.triallength    = 2;  % duration in seconds
> cfg.trialdef.ntrials        = inf;  % number of trials, inf results  
> in as many as possible
> cfg                         = ft_definetrial(cfg);
>
> % preprocess the data
> cfg.channel                 = {'MEG','-TRIGGER','-RESPONSE',...
>    '-UACurrent','-MxA','-MyA','-MzA','-GxxA','-GyxA','-GyyA',...
>    '-GzyA','-GzxA','-MxaA','-MyaA','-MzaA','-E1','-E2','-E3','- 
> E4'}; %read all MEG channels except those with '-'
> cfg.lpfilter                = 'yes';
> cfg.lpfreq                  = 200;
> cfg.bsfilter                = 'yes'; %60 Hz noise
> cfg.bsfreq                  = [58 62];
>
> data = ft_preprocessing(cfg);
>
> In the end, when I go to plot them all on the same graph, the head  
> model seems to be 100 times larger than the gradiometer sensors  
> (they appear as a small ball in the middle of the head). I've tried  
> adjusting the grad.unit parameter from 'm' to 'cm', but it doesn't  
> help the downstream frequency, leadfield, and source analysis.
>
> Can someone tell me how to ensure the gradiometer and mri/grid are  
> on the same scale so I can be sure I have the right model for my  
> beamforming?
>
> Thank you,
> -Allison
> _______________________________________________
> 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: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20110211/d67ac9a1/attachment-0001.html>


More information about the fieldtrip mailing list