EEG & lcmv

[Nathan Weisz]

Hi,

I am struggling with the sourceanalysis of EEG data which was  
recorded with Biosemi 128 electrodes. the data are visual evoked  
potentials. the topographies look very good and dipolar. below some  
code attached.

one thing i find suspicious is that the 'avg.pow' field after  
sourceanalysis contains some negative values ... i reckon this  
shouldn't be the case.

the call to sourceanalysis is very close to calls i did with MEG data  
yielding good localizations.

the greatest potential error source i see is that i use electrode  
positions from the standard_1005 file in order to use the standard  
BEM model. this is done by choosing the index from the standard-1005  
closest to the biosemi elecrode locations (placed on a sphere). for  
this i use matlabs dsearchn function.
mistakes at this step would seriously screw up the forward model.  
however checking the electrode locations on the standard MRI using  
sourceplot looks sensible however ... so at this stage i am not sure  
whether it's a headmodelproblem or something else i am missing /  
overlooked.

as always, any help is greatly appreciated.

cheers,
n

%first search for electrodes closest in the standard 1005
elec4grid=data.elec;
elec4grid.pnt=[-data.elec.pnt(:,2),data.elec.pnt(:,1),data.elec.pnt(:, 
3)]*85;

stelec=read_fcdc_elec('standard_1005.elc');

ind=zeros(1,128);

for i=1:128
ind(i)=dsearchn(stelec.pnt,elec4grid.pnt(i,:));
stelec.pnt(ind(i),:)=NaN; %avoids double choosing of electrodes
end

stelec=read_fcdc_elec('standard_1005.elc');
elec4grid.pnt=stelec.pnt(ind,:);


cfg=[];
cfg.grid=[];
cfg.grid.xgrid='auto';
cfg.grid.ygrid='auto';
cfg.grid.zgrid='auto';
cfg.grid.resolution=10;
cfg.vol=vol; %this is the standard BEM from Fieldtrip / eeglab
cfg.elec=elec4grid;
grid=prepare_leadfield(cfg);

cfg=[];
cfg.covariance='yes';
cfg.preproc.reref='yes';%it should be average reference already. just  
to be sure.
cfg.preproc.refchannel='all';


cfg.latency=[-.3 -.1];
cfg.covariancewindow = [-.3 .1];
baseERP=timelockanalysis(cfg,data);

cfg.latency=[0.1 .3];
cfg.covariancewindow = [.1 .3];
visuERP=timelockanalysis(cfg,data);

cfg=[];
cfg.method='lcmv';
cfg.vol=vol;
cfg.elec=elec4grid;
cfg.grid=grid;

baseS=sourceanalysis(cfg,baseERP);
visuS=sourceanalysis(cfg,visuERP);

visuSn=visuS;
motorSn=motorS;
visuSn.avg.pow=visuS.avg.pow./baseS.avg.noise;
motorSn.avg.pow=motorS.avg.pow./baseS.avg.pow;

%%%COMMENT 17:
%%%The solutions then have to be interpolated on the standard MRI.
cfg = [];
cfg.downsample = 2;
cfg.sourceunits='mm';
visuS_int = sourceinterpolate(cfg, visuSn, mri);
motorS_int = sourceinterpolate(cfg, motorSn, mri);






--------------------------------
Dr. Nathan Weisz

INSERM - Unité 821
Dynamique cérébrale et cognition
Centre Hospitalier Le Vinatier, Bâtiment 452
95 Boulevard Pinel
69500 Bron, France

Tel: ++33 - (0)4 - 7213 8915
Email: weisz at lyon.inserm.fr
Chat-AV: nathanweisz at mac.com
Homepage: http://web.mac.com/nathanweisz
Neurotree: http://neurotree.org/neurotree/tree.php?pid=8692

Please avoid sending me Word or PowerPoint attachments.
See http://www.gnu.org/philosophy/no-word-attachments.html




----------------------------------
The aim of this list is to facilitate the discussion between users of the FieldTrip  toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20070821/8e1fe082/attachment-0001.html>