Hi Wayne,
yes, the problem is that the surfaces are including some non-cortical regions, which makes the topology dramatically wrong, so the topology fixer takes forever. To fix it you have to figure out why they are connected and remove the correction (possibly by erasing voxels in the wm.mgz). If you can't figure this out you can drop us the data and we'll take a look
cheers, Bruce
On Thu, 7 May 2009, Wayne Su wrote:
I closely looked at the two scans having unexpected long processing time. Both them the ribbon.mgz covered not only brain but also outside the brain. One scan the lh.ribbon.mgz only covered a little bit of the cortex area. Is there any way to fix this issue?
Thanks.
Wayne
On 05/05/09 9:48 AM, "Wayne Su" wsu@interchange.ubc.ca wrote:
I re-ran over 400 scans by using Freesurfer 4.3.0 on Mac OS 10.5.6. Well, I¹m very happy on the results. Anyway there are two issues I encountered:
- I have 3 scans hanging on the last second step when creating the
aparc+aseg.mgz by using mri_aparc2aseg (more precisely is the ³labeling slice² processing) for over one week though it finally finished. The log is attached. 2) I have some scans reported ³mris_fix_topology failed! (euler number != 2)² so recon-all exited with errors. I didn¹t find there was any problem when I used the V4.0.5. I attached the log as ³recon-all2.log².
Thanks.
Wayne