Hi Luis, yes there is a more efficient way. Specify the native res functional as the --mov argument and specify the registration that you used to create the hires_functional as the --reg argument (instead of --regheader). While you can use a .w file, I would recommend instead a .mgh file (we're trying to phase out the .w); If you use mgh format, not need to spec a --out_type. That command will convert to the native subject surface. If you want to go to fsaverage you can either run mri_surf2surf after that or spec --trgsubject fsaverage doug
On 06/25/2012 03:43 PM, Luis R. Piloto wrote:
Hello Fellow Freesurfers,
I need some verification that my method for mapping functional data onto a spherical surface is sound.
What I'm currently doing is running recon-all on my high-resolution, structural volume (.nii.gz format). After that, I take a single timepoint of my functional data (that has been both registered and upsampled to the high-res structural space), and run it through mri_vol2surf with the following command: mri_vol2surf --mov highres_functional_slice00.niii.gz --hemi lh --regheader my_subject_name --out_type paint -o ./my_output_surface.w
What I'm most worried about is whether this method of registration (i.e. specifying --regheader) is sufficient to go from the upsampled/registered functional to the surface? Also is there some more efficient way of doing this? I feel like there might be a smarter way to go from the functional to the structural without having to explicitly register and upsample before using mri_vol2surf.
After this step, I simply pass my_output_surface.w to a program that reads the corresponding vertex and functional activity at that location and look up the vertex coordinates in my lh.sphere file. Any feedback would be really appreciated as I'm still really new to this software suite and was given a highly specific task to take care of. Thanks in advance!
Regards, Luis Piloto
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer