[FieldTrip] Forward solution using concentric spheres and BEM models

[Schönherr, Margit]

Hello Juan Pablo,

I am using fieldtrip-20110603.
Now that you say, that your problem persisted, I remember having changed the code of prepare_mesh_segmentation. You find the function in the private directory. In line 25, it sets cfg.tissue = 1 if mri has fields 'brain' or 'scalp'. I think this explains why the segmentation is done for compartment 1 only. So I have simply commented this line, so that cfg.tissue is left unchanged, and the segmentation is done for all compartments.

I know that this is bad style, but I wanted to get it working. And since I was relatively new to fieldtrip, I hesitated to ask the discussion list...

I hope this helps. And I'm open for better solutions.
Best,
Margit


________________________________________
Von: fieldtrip-bounces at donders.ru.nl [fieldtrip-bounces at donders.ru.nl] im Auftrag von Juan Pablo Neira [jpnv2006 at gmail.com]
Gesendet: Freitag, 19. August 2011 16:01
An: Email discussion list for the FieldTrip project
Betreff: Re: [FieldTrip] Forward solution using concentric spheres and BEM      models

Thank you Margit for your help,

I tried your sugestion to create the geometrical description of the
brain, skull and scalp but
I am still having the same output bnd = 1x1 struct (pnt and tri just
of the brain).  I should have
bnd = 1x3 struct (pnt and tri of the brain, skull and scalp).

This is the information in the command line in matlab
not downsampling brain
not downsampling scalp
not downsampling skull
using the segmentation approach
using the segmented MRI
triangulating the boundary of compartment 1
segmentation compartment 1 of 1 completed

Can you tell me which version of fieldtrip (fieldtrip-yyyymmdd) are
you using so i could compare the code of the
function ft_prepare_mesh.

Regards,

Juan Pablo

2011/8/18 Schönherr, Margit <Margit.Schoenherr at uk-erlangen.de>:
> Hello,
>
> I probably cannot help you with all your questions, but currently I'm also dealing with MRI segmentation and volume conductors, so maybe my solutions can be useful for you.
> Concerning your point 5 (geometrical description of the brain, skull, and scalp), I have done it the following way:
>
> cfg             = [];
> cfg.method      = 'segmentation';
> cfg.tissue      = [1 2 3]; % scalp = 1; skull = 2; brain = 3;
> cfg.numvertices = [2000 1000 800];
> cfg.sourceunits = 'mm';
> cfg.mriunits    = 'mm';
> bnd = ft_prepare_mesh(cfg, mri_segment);
>
> I think you have to specify a vector 'cfg.numvertices', a number of vertices for each compartment.
> By the way, at the beginning of ft_prepare_mesh the cfg-field 'numcompartments' is removed from the cfg structure (at least in my fieldtrip version, which is not the latest).
>
> Another question that comes to me is, why you need the segmented MRI for a concentric spheres volume conductor? So, regarding point 6, you can build such a vol-structure for example like this:
>
> vol.unit = 'm';
> vol.o = origin;
> vol.r = [R_brain R_skull R_skin];
> vol.c = [1 1/80 1];
>
> By this, you can also define the radius as you like.
>
> Best,
> Margit
>
>
>
>
> ________________________________________
> Von: fieldtrip-bounces at donders.ru.nl [fieldtrip-bounces at donders.ru.nl] im Auftrag von Juan Pablo Neira [jpnv2006 at gmail.com]
> Gesendet: Mittwoch, 17. August 2011 14:58
> An: fieldtrip at donders.ru.nl
> Betreff: [FieldTrip] Forward solution using concentric spheres and BEM models
>
> Hello,
>
> I am working with an individual MRI and EEG data to do the forward
> solution with two different volume's model (concentric spheres and
> BEM),
> so at the end i can compare the results.  But i did not have the
> expected results of the concentric spheres model until now.
>
> This is my script:
>
> 1.  Read individual MRI
> mri=ft_read_mri('data_patient_1\******.hdr');
>
> %I got this "Warning: flipping 1st dimension (L-R) to obtain volume in
> neurological convention"
>
> %How can i do a homogenous transformation matrix, using the voxel dimensions
> %that are specified in hdr.dime.pixdim?
>
> 2. Realign to 'head coordinates'
>
> 3. MRI Reslicing
>
> 4. MRI segmentation: (brain, skull,scalp)
>
> 5. Create geometrical description of the brain, skull and scalp (3 compartments)
>
> cfg                 = [];
> cfg.interactive     = 'no';     %segmentation method
> cfg.numvertices     = 3000;
> cfg.sourceunits     = 'mm';
> cfg.downsample      = 2;
> cfg.numcompartments = 3;
> cfg.tissue          = {'brain' 'skull' 'scalp'};
> cfg.smooth          = 'yes';
> bnd                 = ft_prepare_mesh(cfg, mri_segment);
>
> %The output just give me the geometrical description of the brain.
> How can i get the geometrical description of
> %the skull and scalp also to proceed to create a 3 spheres concentric
> sphere model?
>
> 6.  If i have the 3 geometrical descriptions (brain, skull and scalp).
>  Create a 3 concentric spheres model.
>
> cfg                 = [];
> cfg.headshape       = [bnd1 bnd2 bnd3];   % brain- skull - scalp
> cfg.conductivity    = [0.33 0.0042 0.331];
> [vol, cfg]          = prepare_concentricspheres(cfg);
>
> %I got the model but part of the brain is outside of its sphere and
> also the skull.  How can I increase the radius so
> % the whole brain will be inside its sphere and also the skull?  The
> sphere of the scalp fix good.
>
> I hope someone can help me solve these questions.
>
> Regards,
>
> Juan Pablo
>
> _______________________________________________
> 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