Hi Bruce,
As you asked, I uploaded the recon-all log-file for that subject, edited and unedited. We're using version 5.3. How can you check for sure that the WM significantly gets >110?
I also uploaded the pictures once again.
Thanks,
Cleo ________________________________________ Van: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu namens Bruce Fischl fischl@nmr.mgh.harvard.edu Verzonden: vrijdag 21 oktober 2016 02:14 Aan: Freesurfer support list Onderwerp: Re: [Freesurfer] Manual pial and white matter corrections cause irregularities
Hi Cleo
can you upload a subject before and after editing so we can replicate what you are seeing? And what version of FreeSurfer are you using? Does the WM after editing get signifcantly >110? If so, you may have accidentally placed a control point in a gray matter voxel
cheers Bruce
On Thu, 20 Oct 2016, Cleo Vandervost wrote:
Hello Freesurfer team,
For my thesis I'm researching the differences in grey and white matter and cortical thickness between children with and without dyslexia. My supervisor advised me to manually edit the Freesurfer-generated data. I've been editing my data as follows:
adding control points to (larger) WM-areas that FS did not include
manually editing (smaller) WM-areas through recon-editing
correcting the pial surface through recon-editing
After finishing and saving my edits I reran recon-all with this command: recon-all -autorecon2-cp -autorecon -subjid (subject number). When I opened my data again in Freesurfer to check my edits, I saw that there were two big problems:
'distortions' in the pial surface
sometimes large areas of WM suddenly were excluded, even when I didn't make any edits in that area
It makes the edited data look much worse than the original data, which is weird I suppose.
I've included some pictures so you can see what I mean.
Picture (a) is the original unedited version.
Picture (b) is the edited version: control points and WM edits. You can see that there's a large area
in the left hemisfere excluded from the original WM area
- Picture (c) is the re-edited version of (b), control points added in that large area. This time it's
included but the pial surface shows a lot of irregularities.
Me and my supervisor tried a couple of times with other data, but we still had the same results. We also tried by only adding control points and editing manually, without the pial-corrections.
Do you have any idea of what's going wrong in the editing proces?
Thank you very much,
Kind regards
Cleo Vandervost
1st master speech and language therapy
Katholieke Universiteit Leuven