Dear Christian,
I'm not the ultimate expert on that but I doubt it's possible to
create a cortical mesh on a regular grid. The whole point of a mesh is
that the vertices can be located in such a way that the local surface
morphology will be captured. So you can either have a surface mesh or
a regular grid which is constrained to grey matter. You can make a
canonical grid by defining it in MNI template space and then applying
deformations to get it to fit your subjects, but then again those
subject-specific grids will not be equidistant.
See
http://fieldtrip.fcdonders.nl/example/create_single-subject_grids_in_individual_head_space_that_are_all_aligned_in_mni_space
and the code of spm_swarp and spm_eeg_inv_mesh
Best,
Vladimir
On Thu, Apr 14, 2011 at 4:09 PM, Cristian Carmeli
<[log in to unmask]> wrote:
> Dear Vladimir Litvak
>
> I am using SPM8 to compute lead-field matrix for some source analysis. To
> fully exploit the inverse methods I am applying, I would need a (canonical)
> cortical mesh defined on a regular grid, that is adjacent points or mesh
> vertices are always separated by the same distance.
>
> In current SPM release, the three canonical cortical meshes available of
> 5124, 8196 and 20484 points do not match that requirement. Is there any
> possibility to have or obtain such a regular mesh?
>
> thank you in advance for your attention
>
> Yours sincerely
>
> Cristian Carmeli
>
>
|