Hi Scott
the flattening is a nonlinear process, so there is no such T that works for all vertices. You can use the registration (epi to structural) matrix from bbregister to compute T for taking EPI to structural coords, then find the vertex that is closest to that voxel and lookup the flattened coords to account for the nonlinearity
cheers Bruce
On Tue, 3 Jan 2012, warre295@umn.edu wrote:
Hello gurus,
Thank you in advance for any assistance. I have seen questions similar to this in the archives, but I believe this is a distinct problem. I apologize if this has been addressed previously.
I am trying to obtain a retinotopic map, but for reasons of compatibility and future analysis I would like to keep as much of the work in MATLAB as is possible. To date, I have used FreeSurfer to generate a flattened cortical surface with great success. I would like to at this point exit the FreeSurfer world and perform the remaining analysis in MATLAB.
What would be most ideal is if I could obtain a matrix which I can import into MATLAB and calculate a statistic for a point on the *flattened* cortical surface as a function of that statistic calculated for each voxel. This includes the phase information for a retinotopic map, but really I am looking for a solution to a more general problem. In other words, I need the matrix T in the equation...
T*[Statistic from each voxel] = [Statistic for each point of Flattened Surface]
where T describes how each point on the surface is derived from some set of voxels.
I imagine this is an incredibly naive question, and that this would require a fair amount of work involving multiple transformations if it is even possible at all. I appreciate any and all advice.
Best wishes,
Scott Warren Graduate Student, Department of Neuroscience University of Minnesota
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer