Hi Bruce,
Yes, this was with the longitudinal stream. The command I used was:
recon-all -base 2829 -tp 20648 -tp 20995 -all
I've uploaded these data to the ftp site: transfer/incoming/RLM_umn_problem_caseB.tgz
tp1 20648 & tp2 20995 were fully processed using the cross-sectional stream (recon-all -subjid ${sub} -autorecon-all).
tp1 is a Siemens Trio va25 8channel head coil acquisition. tp2 is a Siemens Trio TIM vb15 12channel head coil acquisition on the same subject with identical scan parameters.
The problem is with the base subject, 2829.
Also, the bugr report, SuSe-Release & cpuinfo are all in the tar file that was uploaded.
Let me know if you need any additional info, and thanks again for looking into this for us!
Best,
Ryan
On Sep 10 2009, Bruce Fischl wrote:
Hi Ryan,
is this only in the longitudinal stream? Sounds like a bug. If you upload the whole subject directory we'll take a look.
Bruce On Thu, 10 Sep 2009 muet0005@umn.edu wrote:
Hello,
I am running the longitudinal processing using version 4.5 and was wondering if there are any data quality checks specific to the longitudinal stream (other than the standard ones listed for the cross-sectional analysis)? For example, is there a way to check the registration between TP1 & TP2 when creating the base?
Also, in two of my 36 subjects (one in creating the base, one TP->base longitudinal processing), the final step that creates wmparc.mgz was taking extremely long to finish (I let it go 18hrs before killing the process, and only 25 slices were complete). After looking at the aparc+aseg.mgz file, I noticed that part of the right hemi, and all voxels outside of the brain received a value of 41, corresponding to right cerebral white matter. The aseg.mgz appears to be fine, and the pial & inflated + annotation file appeared OK in tksurfer. Any ideas on what may be going wrong with these cases?
Thanks! -Ryan
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer