On 8/13/2020 10:15 AM, Ellen Ji wrote:
External Email - Use Caution
Thanks Douglas.
recon-all was performed using FS V6. However, we have since upgraded to V7. Should I stick with what was originally used (mri_aparc2aseg)?
I don't think it will make much of a difference, so do whatever is easier for you
Here is how the wmparc.mgz was created. Could you verify if my modified steps below make sense?
wmparc.mgz:
mri_aparc2aseg \ --s subject1 \ --labelwm \ --hypo-as-wm \ --rip-unknown \ --volmask \ --o mri/wmparc.mgz \ --ctxseg aparc+aseg.mgz \
modification for my annotation:
The reason why I added wmparc-dmax 2 is because I want to extend the annotation, which only includes gm, into the wm. However, when I do this, there are 2x the labels (gm and wm). Is there any way to do an extension while maintaining 318 regions? (if I remove the labels post hoc using mri_binarize, I believe that would just erase all wm, which isn't what I want).
mri_aparc2aseg \ --s subject1 \ --annot 500_sym.aparc \ --wmparc-dmax 2 \ --labelwm \ --hypo-as-wm \ --o subject1/mri/aparc.500+2mm.nii.gz
Why are you not including --rip-unknown and --volmask? The number of regions is as expected because it labels cortex as well as the adjacent WM, but the adjacent WM gets a different segmentation index. You can merge them back together afterwards if you want.
best,
Ellen
On 8/13/2020 3:39 PM, Douglas N. Greve wrote:
You can look in the recon-all.log file to see how wmparc.mgz is created. Which version of FS are you using? If V6, then the command will be mri_aparc2aseg. If V7, then it will mri_surf2volseg. Either way, modify the command to use your annotation.
On 8/12/2020 5:18 AM, Ellen Ji wrote:
External Email - Use Caution
Dear Freesurfer experts,
I have a surface parcellation (annotation) with 318 labels; see attachments for each hemi. I wish to extract volume-based features (FA and MD) corresponding to these 318 labels. I believe I should perform something like surf2vol to extend the labels into FA space. However, my surface parcellation is not a surface overlay (which I need for --so). Any other ways to do this?
I wish for the output to be a single volume-based parcellation of 318 labels, corresponding to the two surface parcellation annotation files. I will then use this output to get FA measures for each of the 318 regions.
Thank you, Ellen
Ellen Ji, PhD Postdoctoral Research Fellow Psychiatric University Hospital University of Zürich ellen.ji@bli.uzh.ch homanlab.github.io/ellen/
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