Hi,
I have been running recon1 recently and I have noticed that for some of my participants big chunks are missing in the brainmask.mgz. Out of 20 participants, five participants have this problem. The pattern is similar across participants: the missing parts start around slice 70-50 at the bottom (cerebellum) and then move upward. The data can be found in:
/autofs/space/boggart_001/users/oosterwi/construct_struc
It concerns constr011 (which I corrected using the clone method), and constr012, constr016, constr019 and constr020 (which still show the problem).
I think it can be solved by cloning the T1 (aux volume) as is explained in the Freesurfer tutorial. Still, I wonder whether this is the best solution, and why this is happening in the first place.
I appreciate any help or feedback on this. Thanks in advance! kind regards, Suzanne Oosterwijk
Suzanne, You could try running different watershed thresholds to see if it fixes the problem without having to clone it in: http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/SkullStripFix
Allison
freesurfer@nmr.mgh.harvard.edu