I want to bring your attention to a key difference between the "new" reconstruction and the "old" reconstruction. The new recon automatically fills in the ventricles and basal ganglia prior to creating the surface. This creates a different folding pattern, and so the surface registration target has also changed.
The implication is that you should not mix data from "old" and "new" reconstructions in a group analysis. Also, those using the "new" recon should render group results on the new average subject (fsaverage). Those using the "old" recon should continue using average7.
I put "old" and "new" in quotes because a lot of people were using the "new" recon (ie, recon-all-dev) prior to the release.
If you are not sure which atlas you used, then try this:
cd yoursubject/scripts grep GCS recon-all.log
This may produce several lines. Look at the last of the lines for something like:
GCS atlas2002_simple.gcs -----> This means it is the old atlas GCS curvature.buckner40.filled.desikan_killiany.gcs ----> New
Thanks to Frida Polli and Eric Feczko for drawing my attention to this issue.
doug
Nick Schmansky wrote:
Freesurfer community,
A new stable release, v3.0, is now available. The Release Notes page describing what's new, and a link to the download page, is here:
http://surfer.nmr.mgh.harvard.edu/fswiki/ReleaseNotes
All users, including those using prior 'dev' releases, are encouraged to upgrade to this release.
Updates to the stable release will be issued if major bugs are found (and fixed), and as new features which do not affect existing functionality become available.
The Freesurfer development team
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer