Hello,
I'm trying to resample Freesurfer output volumes and surfaces back into their native anatomical space using mri_vol2vol and mri_surf2surf. The volumes appear to be resampled correctly, their resolution and orientation match the original images'.
However, the resampled surfaces do not appear to have been transformed correctly by mri_surf2surf. When I visually compare them to the Freesurfer-space surfaces overlaid on the brain images in FreeView, TKMEdit, Paraview-CAWorks, etc., they seem to be identical to the un-transformed, Freesurfer-space surfaces. In addition, when I check the list of vertices for transformed and untransformed surfaces, the coordinates of each vertex are unchanged.
These are the commands that I'm using:
mri_vol2vol --mov brain.mgz --targ rawavg.mgz --regheader --o resamp_brain.mgz
mri_surf2surf --sval-xyz pial --reg resamp_brain.mgz.reg --tval resamp_lh_pial --tval-xyz --hemi lh --s 1158119_3_6
The registration file, "resamp_brain.mgz.reg" looks like this:
subject-unknown
0.937500
0.937500
1.000000
1.000000000000000e+00 0.000000000000000e+00 0.000000000000000e+00 0.000000000000000e+00
0.000000000000000e+00 1.000000000000000e+00 0.000000000000000e+00 0.000000000000000e+00
0.000000000000000e+00 0.000000000000000e+00 1.000000000000000e+00 0.000000000000000e+00
0 0 0 1
round
My Freesurfer version: freesurfer-Linux-centos4_x86_64-stable-pub-v5.1.0
Could it be that mri_surf2surf is not applying scaling transformations correctly?
Thanks for your help,
Geoff