Hi Freesurfers, I am a little stuck. I have some 3T data that has had the skull stripping manually corrected. In addition we have some unusual bias artifacts that FAST seems to handle better than n3, so I have done my own bias correction. Since I had to run FAST anyway, I created brainmask.mgz from the partial volumes I have. So I did an mri_convert --conform on both to get nu.mgz(which is skull stripped) and brainmask.mgz.

So now I need to deal with the transformation issues. I was thinking I could just run recon-all -gcareg to get talairach.lta and then copy that to talairach_with_skull.lta and clip out the transform matrix and substitute it into a copy of talairach.xfm. But none of this seems to work. So if anyone has any suggestions I would greatly appreciate them, also I may be missing something, so if there is another step or file I am missing before I can run autorecon2 and autorecon3 please enlighten me.

Thanks in advance, -Mira

Mira Michelle Raman Center for Interdisciplinary Brain Sciences Research Stanford University