Prapti,

Also beginning with version 4.5.0, the random number generator is at a fixed seed value, so re-running will result in exactly the same surface. For this particular lGI problem, you can try adding the flag -randomness to the end of the recon-all stream with -autorecon2 -autorecon3, and a slightly different surface vertex combination will result, which might allow -localgi to succeed.

Nick

On Thu, 2009-10-22 at 12:54 +0200, Marie Schaer wrote:

Dear Prapti,

Which version of freesurfer do you use? Have you carefully checked that your surfaces are anatomically correct? (i.e. follow accurately the cortical mantle without gross abnormality)

Marie

Quoting Prapti Gautam Prapti.Gautam@anu.edu.au:

...

Dear freesurfers,

I have a problem running localGI in one subject. I previously got an error message that said that one of the vertex value was aberrantly high. The mris euler number on the subject was fine - but I relaunched the recon-all -autorecon2 -autorecon3 for the hemisphere in question and tried re-running the localGI command since this has been previously suggested as a fix to the problem. However, I get the same message (aberrantly high lGI value) even after re-running autorecon2 and autorecon3. Is there anything else I can try to complete lGI successfully?

Thanks for your help,

Prapti

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer