Did you check the skull strip? You can send it to us if you want and we will take a look.
On Jun 15, 2009, at 10:43 AM, Deane Aikins deane.aikins@yale.edu wrote:
Hi Bruce-
Agreed. Visual check of talairach looks good and manual correction of slight sagital angle fails to improve z value, hence the current notal-check.
Looking at recon-all.log, there is a statement that The surface validation has detected a possible Error", with a suggestion to use the '-atlas' option if the final seg is not valid. Also, it's saying that brainmask.mgz already exists and that I would need to add -clean-bm to overwrite brainmask.mgz. \n
Is any of that information of use?
Deane
Quoting Bruce Fischl fischl@nmr.mgh.harvard.edu:
Hi Deanne,
something is badly wrong - that step should only take a couple of minutes. Did you check the talairach visually? It may need to be manually corrected.
cheers Bruce On Fri, 12 Jun 2009, Deane Aikins wrote:
I'm rerunning it with -no-talcheck. It looks to be hanging on finding
control
points in the Left_Cerebral_White_matter. Can this one step take more than
24
hrs? The "usual" time for a recon-all was @30 hrs for the other brains.
Thanks,
Deane
Quoting Nick Schmansky nicks@nmr.mgh.harvard.edu:
Deane,
the -no-talcheck flag is your best option now, considering that a manual alignment is no better than the automatic.
What is the age range of your subjects? We're implementing an alternate target consisting of young adults scanned at 3T, which works better for that cohort over the adults in the default target we use.
Nick
On Fri, 2009-06-12 at 09:43 -0400, Deane Aikins wrote:
Hello,
I had a Talairach QA check failure, with the error log indicating a Z =
-8
which
was below the criterion of z =-6.
I then used tkregister2 to attempt a manual alignment (no surf orig was available). After visual inspection, the subject and Talairach volumes
looked
really closely aligned already. I did some very conservative rotation
and
movement in the sagital plane, checked other planes, hit "SAVE REG" and
then
executed
recon-all -all -subjid MYFILENAMEHERE
in the subj directory. I then get the same error, with the same z score.
I've
tried several alignments (very little adjustment each time) with the same results.
I notice that when I hit SAVE REG there is a new 4 x 4 numerical matrix
written
to my xterm screen. The values look very very similar to those prior.
Is there a way to interpret these values to assist me in my manual
alignment?
Should I just bag it and use -notal-check? The majority of my patient
set
ran
without errors.
Thank you!
Deane Aikins, PhD Director, Cognitive Neuroscience Laboratory National Center for PTSD Clinical Neuroscience Division Assistant Professor, Department of Psychiatry Yale School of Medicine _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer