Dear all Freesurfer users, I have a few questions regarding wm.mgz edits. I am very new to this area and have only recently started using Freesurfer so please excuse me if my questions appear naive or misinformed. I have managed to make most of the necessary edits needed to fix skullstrip errors, intensity normalization errors and pial errors, but I am a bit stuck when it comes to fixing topological defects. My question relates to what should or should not be included in the wm.mgz. Apart from the optic chiasm, which I have removed, is it normal to have, what appear to be, multiple specs included in the wm.mgz which when the aseg is loaded turn out to be parts of the amygdala and hippocampus. When I view the smoothwm.nofix these appear as jagged shard like shapes protruding out from the surface. In one publication I saw recently it was suggested that these should be removed but should this be done if they are parts of the amygdala and hippocampus. I am worried about this as these are the 2 main subcortical areas I will be focusing on. Also, is it normal for the wm.mgz to include parts of the cerebral cortex (labelled as such in the aseg). Also, most of the scans I am working with have numerous holes and handles and this is taking a very long time to fix. I have tried to run the automatic topology fixer again. This has helped a bit but the scans still require a lot of edits. Could this be down to the quality of the scans or could it be that I have edited too much - control points for example. I have followed the advice given and have tried not to make too many edits, but having said that it has taken me a very long time. Any advice in relation to any of these questions would be very very very much appreciated. Best wishes fuzzy