dataset-audit

You can audit a medical imaging dataset by running this Skill on a local directory or archive to read actual file headers and cross-reference official documentation, producing a ground-truth report of modalities, dimensions, and subject counts.

Yes, this dataset audit tool supports 2D, 3D, and 3D+t data across MRI, CT, and US modalities, verifying dimensions, views, temporal frames, and segmentations to ensure accurate interpretation before publication.

You can validate ground-truth segmentations and meshes by inspecting contour data and cross-referencing official dataset documentation, ensuring available annotations and data types are accurately reported before starting a study.

Yes, you need specific Python dependencies including pydicom, nibabel, numpy, and trimesh to read actual DICOM and NIfTI file headers and inspect mesh data during the dataset auditing process.

Your medical imaging dataset audit may show different dimensions because this Skill reads actual file headers rather than relying on documentation, revealing true on-disk contents to prevent misinterpretations and ensure data governance.