[Eeglablist] Head model

[Robert Oostenveld]

Dear Jose,

No, it is not a precomputed leadfield. It is a matrix that describes  
how the infinite medium potential at all 3000 vertices (i.e. the  
potential if there were no conductivitu differences between brain- 
skull-skin-air) should be corrected, to account for the conductivity  
discontinuities at the three boundaries. This matrix is multiplied  
with the infinite medium leadfield on all 3000 vertices to compute  
the "bounded" potential distribution on all 3000 vertices.  
Subsequently, the potential on the outer 1000 vertices is linearly  
interpolated onto the ~128 electrodes (these two steps are done at  
once for efficiency).

The formulation of the boundary element method allows for seperating  
the volume conduction (geometry+conductivity) terms from the source  
terms: the "mat" matrix contains only volume conduction aspects, and  
depending on how you compute the leadfield (i.e. one dipole, or a  
distributed patch) the infinite medium leadfield only contains the  
source term.

The actual code in Fieldtrip is not so transparent, but that is  
because of speed considerations. If you want to know how it is done,  
you should look through the fieldtrip/private/prepare_vol_sens  
(computes the linear interpolation from outer surface vertices  
towards electrodes) and fieldtrip/private/compute_leadfield (computes  
the infinite medium potential on all vertices and then the bounded  
medium leadfield on all electrodes).

best regards,
Robert

On 5 Sep 2006, at 10:24, Jose Rebola wrote:


> Dear All,
>
> Can anybody tell me what exactly is the meaning of the field mat of  
> the structured vol in the BEM example for the head models in  
> EEGLAB? It is a 3000x3000 double. Is it a pre-computed leadfield?  
> And if so, how is it ordered?
>
> Thanks in advance,
>
> José Rebola
> _______________________________________________
> eeglablist mailing list eeglablist at sccn.ucsd.edu
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist- 
> unsubscribe at sccn.ucsd.edu
>