Hi Catherine, If it's the mtl pial surface around the hippocampus then the pial surface is not expected to be accurate in that region. See FAQ question at link below.
http://surfer.nmr.mgh.harvard.edu/fswiki/UserContributions/FAQ#Q.Thesurfaces...
-Louis
On Wed, 9 Jan 2013, Catherine Bois wrote:
Hi,
I am currently editing some scans (ie adding control points etc), however I keep finding that after the editing often the brains look "worse" than they did before; especially the medial temporal lobe area. I have taken care not to put too many control points in/made sure not to place them in partial volume areas, and only in areas that are clearly excluded white matter. This problem has happened repeatedly, yet looking at scans that my colleagues have edited (some with up to 2000 control points) this problem does not occur. Has this happened before, is it the software "malfunctioning" in some way?
Thank you very much for your reply,