External Email - Use Caution
Hi Fatih,
Instead of mri_label2vol and mri_concat, you can simply call
mri_aparc2aseg --s yoursubject --annot Schaefer2018_400Parcels_7Networks_order_native --o $SUBJECTS_DIR/yoursubject/mri/outputfile.mgz
One thing to note: "mri_aparc2aseg" will automatically add 1000 for left cerebral cortex voxels and add 2000 for right cerebral cortex voxels. Therefore, for example, a left hemisphere voxel in network 5 will have a value of 1005 in your outputfile.mgz. Here is our suggestion.
We are working on an addition to recon-all, but it won't be so quick :)
Thanks, Thomas
On Fri, Mar 15, 2019 at 4:26 AM Fatih SOĞUKPINAR < fatihsogukpinar96@gmail.com> wrote:
External Email - Use CautionDear FreeSurfer developers *For my MRI data, I have cortical parcellations and segmentations for the Destrieux and Desikan-Killiany atlases. Now, **I am attempting to use Schaefer 2018 atlas (https://github.com/ThomasYeoLab/CBIG/tree/master/stable_projects/brain_parce... https://github.com/ThomasYeoLab/CBIG/tree/master/stable_projects/brain_parcellation/Schaefer2018_LocalGlobal) for cortical parcellation together with FreeSurfer. Currently what I have tried is:*
*1)* mri_surf2surf --srcsubject fsaverage --trgsubject $SubjectID --hemi lh --sval-annot /ourSchaferAtlasDirectory/label/lh.Schaefer2018_400Parcels_7Networks_order.annot --tval ourSubjectDirectory/label/lh.Schaefer2018_400Parcels_7Networks_order_native.annot *2) *the same of 1) for the right hemisphere *3) *mri_label2vol --annot ourSubjectDirectory/label/lh.Schaefer2018_400Parcels_7Networks_order_native.annot --temp ourSubjectDirectory/mri/our T1.nii --o ourSubjectDirectory/mri/sch_lh.mgz --subject $SubjectID --identity --hemi lh
- the same of 3) for the right hemisphere
- 5)* mri_concat ourSubjectDirectory/mri/sch_lh.mgz
ourSubjectDirectory/mri/sch_rh.mgz --sum --o ourSubjectDirectory/mri/sch_lh+rh.mgz
However, at this point, the resulting .mgz file (sch_lh+rh.mgz) is a bit strange : When I open the file in matlab, I see that it's volume field is 218*182*182 , while the same data results in 256*256*256 for Destrieux Atlas. Furthermore, values of the volume field of sch_lh+rh.mgz change between 0-200 , while this is 0-12175 for the Destrieux Atlas. These effects are also apparent in the ultimate results (parcellation.nii images, which I generate by using .mgz files). They are not aligned properly and their intensity is also erroneous. I suspect that we have a mistake in the above pipeline.
*My questions are: *
- Is there simpler way to do this task? (For example, a simple addition
to the recon-all function, etc.)*
- If not, what else should I do to be able to get the correct
parcellations ? * *PS: I am a bit new to the FreeSurfer and MRI processing, so please accept my apologizes if this is a simple question... * Thank you in advance and best regards. _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer