help
help > RE: Importing .annot ROIs
Oct 8, 2015 06:10 PM | Stephen Bailey
RE: Importing .annot ROIs
Aye, Alfonso had already told me the answer to this in the
past. Here is his very thoughtful explanation for anyone else who
is running into this problem:
I believe this error probably indicates that you
are entering subject-specific subject-space .annot files (e.g. the
ones in your subject/label/ folders) instead of subject-independent
fsaverage-space .annot files (the ones in your freesurfer
distribution subjects/fsaverage/label/ folder). Simply select the
latter .annot file instead, and that should work without a
problem.
Just to clarify, for surface-based analyses CONN will be using your subject-specific lh|rh.sphere.reg files (in the subject/surf/ folders) generated by freesurfer to figure out the association between subject-space and fsaverage-space for each subject when extracting cortical-surface functional data. When you enter surface-level ROIs (i.e. .annot files or .img surface-mask files) those ROIs are applied to the surface-level functional data (already extracted cortical-surface functional data), so CONN expects those ROIs to be in fsaverage-space already. When you enter volume-based ROIs (e.g. .mgz or .img/.nii volume-mask files) those ROIs are applied to the volume-level functional data instead (before cortical-surface extraction), so CONN expects those ROIs to be in the same space as your functional data.
Just to clarify, for surface-based analyses CONN will be using your subject-specific lh|rh.sphere.reg files (in the subject/surf/ folders) generated by freesurfer to figure out the association between subject-space and fsaverage-space for each subject when extracting cortical-surface functional data. When you enter surface-level ROIs (i.e. .annot files or .img surface-mask files) those ROIs are applied to the surface-level functional data (already extracted cortical-surface functional data), so CONN expects those ROIs to be in fsaverage-space already. When you enter volume-based ROIs (e.g. .mgz or .img/.nii volume-mask files) those ROIs are applied to the volume-level functional data instead (before cortical-surface extraction), so CONN expects those ROIs to be in the same space as your functional data.
Threaded View
| Title | Author | Date |
|---|---|---|
| Stephen Bailey | Oct 8, 2015 | |
| Stephen Bailey | Oct 8, 2015 | |
| Thinh Nguyen | Nov 30, 2015 | |
| Alfonso Nieto-Castanon | Nov 30, 2015 | |
