Hi Jasmin
the norm.mgz isn't a voxel labeling - it's an intensity volume. We typically check the aseg.mgz display *over* the norm.mgz to see if the labels make sense w.r.t. the intensities.
cheers Bruce On Sat, 2 Jul 2016, Jasmin Alves wrote:
Dear Freesurfer support, I noticed in previous threads, people have inquired about how to check the quality of the aseg.mgz segmentation using the norm.mgz as a reference. When I did this, there were several mislabeled regions according to the norm.mgz ie. voxels labeled as white matter on the norm.mgz but the hippocampus on the aseg.mgz. Other times the accumbens was labeled as the putamen or the pallidum as the amygdala. This was just in one participant, other participants had several more discrepancies. Are these discrepancies something that requires manual editing? Because I feel like I will have to edit almost every voxel. Your suggestions would be greatly appreciated.
Thanks, Jasmin
-- Jasmin AlvesPredoctoral Student Medical Biology Graduate Program University of Southern California jalves@usc.edu