Hi Shanqing, this is a known problem. One way around it is to convert the label to an annotation (mris_label2annot), then run mri_aparc2aseg (converts surface info into a volume), then mri_cor2label to extract the final label. It's not elegant, but it works.

doug

Shanqing Cai wrote:

Hi,

I am using mri_label2vol to generate some volume ROIs from surface labels. I noticed that the results contain "holes", by which I mean voxels that aren't label and are surrounded by labeled voxels. This seems to happen regardless of the step size I specify after the "--proj" option of mri_label2vol and this happens in both the gray- and white-matter projections. I am using FS 5.1.0.

I wonder whether other people have experienced similar problems and how these holes can be gotten rid of.

Thanks!

Best regards, Shanqing Cai _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer