Hi Gabor

don't worry about the wm.mgz. It's an intermediate volume created during the process, but not used for any morphometry (you are correct - we use the white surfaces)

cheers Bruce On Thu, 9 Feb 2012, Gabor Perlaki wrote:

Dear all,

I'm new in freesurfer. I ran "recon-all -all" on 8 subjects without error, and I loaded the wm.mgz and the surfs. White matter surface (yellow line) seems pretty good by visual inspection, but the wm.mgz seems not so good everywhere. There are certain white-matter areas, which are clearly inside the yellow line (WM surface) but they are not considered as white-matter on the wm.mgz image. So it seems for me that calculating white-matter volume is more accurate using the white-matter surface instead of wm.mgz. What does wm.mgz used for? Does it has any impact on the white-matter volume reported by freesurfer? Is it a problem, that wm.mgz doesn't fit well the white-matter surface everywhere? What is the reason that wm.mgz doesn't fit the WM surface everywhere?

Thanks a lot, Gabor