Hi Hakon,
have you checked the brainmask in the longitudinal run? It should be OK, after fixing the base. If it looks OK, then the problems with the pial surface must be due to something else.
You can do edits in the longitudinal run of that timepoint and re-run as you would do in the standard (cross sectional) FreeSurfer stream.
Best, Martin
On Thu, 2009-10-15 at 14:42 +0200, hakon.grydeland@psykologi.uio.no wrote:
Hi Martin,
thanks for your answer!
Editing and re-running the base (as a cross-sectional from after canorm) improved the pial surface of the base, but did unfortunately not improve the pial of the final output (<tpNid>.long.<baseid>). Is it possible/recommended to do the manual edits at other points in the longitudinal stream, e.g. to the final output itself? If so, how would you re-run it?
Thanks again,
Håkon
Hi Hakon,
I have not yet tried this, but I would recommend (see below):
On Mon, 2009-10-12 at 17:10 +0200, hakon.grydeland@psykologi.uio.no wrote:
Hi,
I have a couple of questions regarding manual edits such as deleting non-brain tissue included in the pial surface in the longitudinal stream (v4.5):
- Where in the longitudinal stream would you recommend doing edits? On
the brainmask.mgz of the base?
Yes, fix it in the base, because the brainmask.mgz is mapped from the base to all timepoints and used in the -long stream.
- If so, would you then re-run the base as a cross-sectional (recon-all
-s baseid -autorecon2-pial -autorecon3)?
No, there are differences in the base from the regular cross-sectional stream (e.g. gcareg and canorm are different). Therefore you need to rerun the base stream from that location. I have never tried this, but do not see a reason why it should not work.
Good luck, Martin.
Thanks,
Håkon
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer