Dear freesurfer experts,

I have a question about making edits in the freesurfer results.
During the workshop which I attended this year, I learned about white matter edits and control points edits when white matter is not segmented accurately. I have used these tools quite successfully in our projects.

However, in our research in multiple sclerosis, often times we encounter cases with lesions on grey matter. The lesions in grey matter causes intensity of the voxel over that region to become hypo-intense like CSF. Due to this, the pial surface line is pushed inward into the lesion, missing some grey matter around it. Since, cortical thickness is the most important measure in our research, we could like to be able to solve this issue. We would like to make sure that the pial surface line goes around the lesion without missing grey matter for accurate thickness calculation.

In order to solve this issue we have tried to first delelte and fill the lesion in the brainmask.mgz with voxels with grey matter intensity in the region using the voxel editing tools (instead of 255(default) used in white matter edit, we use 70-75 for grey matter). After editing, we run recon-all with autorecon-2 and autorecon-3.  Quite surprisingly, this strategy has not resolved our problem. My understanding is that the pial and main surface lines (red and yellow) were draw based on the intensity of voxel. Our question is, why would the lesions which are now filled with grey matter not treated as grey matter?

We have tried this strategy, and it has not worked. What would you recommend us for solving such issue?

I have attached a screen shot of a slice where you can notice that in the top right corner, the lesion on the grey matter causes the pial surface line to miss grey matter around it. Ideally, we would like the pial surface line to go around the normal grey matter surface.

Thank you,

Sankalpa Khadka

Research Assistant
Neurolgy, Yale School of Medicine