Freesurfer/recon all: error in segmentation of pial and white


Hi all,

Even though in the “recon all.log”: it says: recon-all -s sub025 finished without errors, after checking the pial and white matter segmentations, very clear misplacement failures and part of dura mater are present (see attachment: segmentation misplacement and dura

I hope someone of you could help me with solving this error.

I am running recon all from the terminal of a linux machine: freesurfer-Linux-centos6_x86_64-stable-pub-v6.0.0-2beb96c.

I tried recon all on a T1 of small vessel disease patients. I tried the command on several subjects and I got the same outcome: a bad allignment of pial and white surfaces.

For sake of curiosity, I tried the recon all command on another dataset (stroke) and with these data I got an hard failure.

recon all exited with errors:

ERROR: talairach_afd: Talairach Transform: transforms/talairach.xfm FAILED (p=0.0573, pval=0.0034 < threshold=0.0050). (see second recon all o35403143 with errors)

I would like to add the recon all log to this post but apparently new users are not allowed to do so.
Any advice is appreciated!
Thank you in advance!