Hello Freesurfer experts,

I would like to follow up on my previous question. As Doug recommended, I did the manual editing to the wm.mgz by deleting voxels in the wm.mgz file, which I thought are gray matter.

I rerun recon-all using the command below because I also did edits to the brain mask (removing some skull) and adding control points.

recon-all -s $1 -all -parallel -openmp 10 -hippocampal -subfields-T1

I do find spots where the editing lead to an improvement, but I have other spots, which did not improve despite wm editing.

And for those spots the new wm.mgz does not label voxels as wm (according to my edits), but the ?h.white would still extents beyond those voxels.

Or the new wm.mgz does label again non wm voxels as wm.

Is there anything else I can try to fix those spots?

Thank you very much in advance for any help!

Carolin

On Tue, Aug 6, 2019 at 8:15 PM Greve, Douglas N.,Ph.D. <DGREVE@mgh.harvard.edu> wrote:

ps, and I meant to say that the non-WM will be labeled as WM.

On 8/6/2019 6:44 PM, Douglas N. Greve wrote:
> Then re-run with
> recon-all -s subject -autorecon2-wm -autorecon3
>
>
> On 8/6/19 6:42 PM, Greve, Douglas N.,Ph.D. wrote:
>> You should edit the wm.mgz. In these places, the WM will be labeled as
>> non-WM. Just erase the mislabeled voxels.
>>
>> On 8/6/19 6:31 PM, Caroline Chwiesko wrote:
>>> External Email - Use Caution
>>>
>>> Hello Freesurfer experts,
>>>
>>> I wrote a previous email about my issue that in many of my subjects I
>>> have spots where the ?
>>> h.white is extending too much into the gray matter after running
>>> recon-all.
>>>
>>> As I understood from a conversation from somebody else,
>>> [[Freesurfer] WM surface including too much gray matter], you are
>>> currently looking into whether default parameter adjustments in
>>> recon-all can fix this problem. Until then, it was recommended to
>>> edit those errors manually.
>>>
>>> My question is what editing technique should I use if I want to
>>> manually edit thesis type of errors? Removing the voxels that I think
>>> are gray matter from the white matter file using recon edit?
>>> I am a little confused about this because I noticed that the ?h.white
>>> boundary does not necessarily follow the wm boundary (see attached
>>> image), so I am wondering, if changes to the wm will fix the problem
>>> that I have.
>>>
>>> Thank you very much in advance!
>>> Carolin
>>>
>>> Screen Shot 2019-08-06 at 1.32.16 PM.png
>>>
>>> _______________________________________________
>>> Freesurfer mailing list
>>> Freesurfer@nmr.mgh.harvard.edu
>>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
>>
>> _______________________________________________
>> Freesurfer mailing list
>> Freesurfer@nmr.mgh.harvard.edu
>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
>

_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer