Hi Alex, does the output of mri_vol2surf look ok on the surface? When you run mri_cor2label, you should use the white or pial surface instead of inflated, but I don't think that is the problem here. Another thing to look at is tkregister2 --mov volumetric-mask --regheader --reg junk.dat --s subjectname and make sure that the label is in the right place (hit the compare button to flip back and forth). doug
On 08/17/2012 11:02 AM, Alex Kell wrote:
hi freesurfers,
i'm trying to move a volume from a subject's native anatomical space to that subject's surface space, and i'm running into some difficulty.
we have binary masks in each subject's native anatomical space and we'd like to move these to be labels in that subject's surface space. there we'd edit them (either manually or with dilation and erosion) and extract some simple anatomical measures (like thickness) from that region in the surface space. to do this, i am using mri_vol2surf to get the volume in the subject's surface space, and then using mri_cor2label to convert this surface file to a label. the problem is that when i look at the label in the surface space, it is in clearly the wrong position -- even though when i look at the volumetric mask as an overlay in tksurfer it looks like it's in the position that it's supposed to be. for instance, see the attached. magenta is the volumetric mask overlay (in the correct position) and the blue-red circle is the label file (in the wrong position).
my calls:
mri_vol2surf --mov<volumetric_mask> --reg identity.dat --hemi rh --inflated --o<output_surf_file> mri_cor2label --i<output_surf_file> --id 1 --l ./<surf_label> --surf <subj> rh inflated
"identity.dat" is a tkregister-style registration matrix where i just put ones on the diagonal and zeros everywhere else. i also tried mri_vol2surf where i used the --regheader flag.
any thoughts? what's going on here?
thanks in advance.
alex
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer