Dear surfers,
I was visually inspecting the results of the auto-recon process using tkmedit (e.g. tkmedit subject brainmask.mgz lh.white -aux T1.mgz -aux-surface rh.white -segmentation aseg.mgz /usr/local/freesurfer/FreeSurferColorLUT.txt)
I noticed that often times the segmentation results from the surface files (pial, main and original surface represented by red, yellow, and green, respectively) are not consistent with the results from the aseg.mgz files. For instance, a piece of gray matter in the anterior temporal pole is colored pink and indicates Left-Cerebral-Cortex as the label, but this area is marked to be outside the red line (pial surface). Likewise, yellow line often cuts through the area indicated as white matter by aseg.mgz.
From a previous email thread, I noted that Bruce mentioned that "We usually
trust the surfaces more than the aseg for cortical morphometry." Does this mean that aseg.mgz is not used in subsequent processing stream? What shall I do if I (after visually inspecting this) want to correct the surfaces more similar to the results given be aseg.mgz?
Thanks for your help in advance.
Joon
-- Joonkoo Park Ph.D. Candidate, Cognition & Cognitive Neuroscience Department of Psychology, University of Michigan