Feedback requested: documenting limits and provenance in an ISLES'22 MONAI notebook

[ringquistchase-collab]

Hello MONAI community — I maintain an independent research repository with an experimental MONAI notebook for ischemic stroke lesion segmentation. This is not a MONAI project.

I’ve updated the notebook/README to describe the work as unvalidated, remove a hard-coded external dataset download, and require a verified, authorized local dataset. The provenance and reuse terms for arrays already included in the repository are still unresolved; I have not run training or reproduced the reported outputs.

If you have experience with MONAI notebooks or research-data documentation, could you offer feedback on whether the docs clearly distinguish assumptions from validated results, and how best to document dataset citation/access terms and provenance gaps? I’m not asking for clinical review or endorsement, model-performance validation, or anyone to access or redistribute data.

PR: ringquistchase-collab/ISLES_2022_FT#1