PDA

View Full Version : BEM modeling


yarikoptic
July 13th, 2005, 23:23
Dear Developers

I came back to brainstorm in attempts to compute BEM for MEG. I took Phantom database, corrected montreal_data/montreal_brainstormstudy.mat so it doesn't choke with window$ backslash notation in the name of BrainStormSubject field, so it started up nicely in Matlab R14. Then I've tried to use
montreal_4layer_tess to compute BEM model, but it chokes after alignment with

Found 2 nodes of surface 3 inside inner surface 2
??? Index exceeds matrix dimensions.

Error in ==> bem_xfer at 553 (error:/home/yoh/rutg/NEM/brainstorm/Toolbox/bem_xfer.m,553,1)
cdvsum=ss(whichsurf);

Error in ==> bst_headmodeler>bem_GainGrid at 2276 (error:/home/yoh/rutg/NEM/brainstorm/Toolbox/bst_headmodeler.m,2276,1)
nfv = bem_xfer(R_eeg,R_meg1,O_meg1,Vertices,Faces,Param( 1).Conductivity,Param(1).mode, ...

Error in ==> bst_headmodeler at 1168 (error:/home/yoh/rutg/NEM/brainstorm/Toolbox/bst_headmodeler.m,1168,1)
BEMGaingridFileName = bem_GainGrid(DataType, OPTIONS, BEMChanNdx); % Computation of the BEM gain matrix on the 3D interpolative grid for MEG and/or EEG data

Error in ==> headmodeler_gui>HRunHeadModeling_Callback at 1282 (error:/home/yoh/rutg/NEM/brainstorm/Toolbox/headmodeler_gui.m,1282,1)
[G,OPTIONS2] = bst_headmodeler(OPTIONS.StudyFile,OPTIONS);

Error in ==> headmodeler_gui at 222 (error:/home/yoh/rutg/NEM/brainstorm/Toolbox/headmodeler_gui.m,222,1)
feval(varargin{:}); % FEVAL switchyard

??? Error while evaluating uicontrol Callback.


Is there anything to make BEM work? am I doing something wrong or do you need more details?

yarikoptic
July 16th, 2005, 05:01
ok - it was kinda my fault -- I've selected too many boundaries for BEM -- I included White Matter as one of the boundaries and here is the problem: I can't find any place in GUI where conductivities are defined for the matters in BEM modeling? That is why the problem occured because by default BST provides an array of 3 values (brain/skull/scalp I believe) as for conductivies if they are not provided during the call to the function. So how can I provide custom conductivities without altering inside code of BST?