You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Once in a while, users will request that we add a new atlas to the list of built-in atlases for XCP-D. Unfortunately, that list is growing quite long at this point. I would prefer to have a solid way to let users provide their own atlases.
The relevant BEP, BEP038: Atlas Specification, proposes adding an "Atlas" dataset type. XCP-D currently writes out its atlases in a manner that somewhat follows the BEP, so I don't think it's a massive stretch to ingress atlas datasets to be used for parcellation.
Additional details
We may need to wait until the Atlas BEP is finalized and merged before XCP-D can officially start supporting it.
Next steps
Allow atlases in --atlases to be specified with the label=path structure, like how fMRIPrep handles --derivatives.
Load the relevant information from the atlas dataset.
In the near term, limit atlases to ones in MNI152NLin6Asym space (for nifti processing) or fsLR+MNI152NLin6Asym space (for cifti processing). In the long term, we should be able to use workflows from fMRIPost pipelines to warp to whatever space is necessary- at least for niftis.
The text was updated successfully, but these errors were encountered:
Summary
Once in a while, users will request that we add a new atlas to the list of built-in atlases for XCP-D. Unfortunately, that list is growing quite long at this point. I would prefer to have a solid way to let users provide their own atlases.
The relevant BEP, BEP038: Atlas Specification, proposes adding an "Atlas" dataset type. XCP-D currently writes out its atlases in a manner that somewhat follows the BEP, so I don't think it's a massive stretch to ingress atlas datasets to be used for parcellation.
Additional details
We may need to wait until the Atlas BEP is finalized and merged before XCP-D can officially start supporting it.
Next steps
--atlases
to be specified with thelabel=path
structure, like how fMRIPrep handles--derivatives
.The text was updated successfully, but these errors were encountered: