Dear all,
we are currently processing a dataset using FreeSurfer 5.0 and in a (very) few subjects (3 of 450) the output is distorted. However, when rerunning the 3 subjects adding the -flirt-motioncor flag to the recon-all -all command, the output is fine (please see http://folk.uio.no/haakongr/files/inflated444.jpg for an example, left side is the -flirt-motioncor version).
Any ideas as how to solve this problem without mixing procedures (or rerunning all subjects) would be highly appreciated!
Thanks,
Håkon Grydeland
---
FREESURFER_HOME: /psyklab/local/freesurfer.5.0.0
Build stamp: freesurfer-Linux-centos5_x86_64-stable-pub-v5.0.0
RedHat release: Red Hat Enterprise Linux Client release 5.5 (Tikanga)
Kernel info: Linux 2.6.18-194.11.4.el5.usit_CVE_2010_3081 x86_64