We are having problems with some of our pial surfaces extending into the sagittal sinus. We are interested in cortical thickness around the precuneus, etc. I attached a couple screen shots to describe. The error is not repaired by adjusting watershed parameters. How do you suggest we proceed? Should we erase voxels from brainmask? We are concerned since these errors extend for a significant number of slices, in many subjects, and include such a large amount of the sinus.

As a second part to this question, I am unclear on how cortical thickness is calculated and whether the statistics directly come from the visible surfaces. For instance, when there is an error with the surface image of the sagittal sinus, it also visibly appears as an error in aseg and aparc. However, the mistaken area doesn't have an associated regional label when we move the cursor over it. Do our cortical thickness statistics reflect the parcellation, which doesn't include the sinus, or the surface, which does include the sinus?

Brianna