AbdomenAtlas 1.0 Mini: abdominal CTs with nine organ masks (first 100 of 5 195 cases)

pl.load("hf_abdomenatlas_mini") loads this dataset. AbdomenAtlas 1.0 Mini (Johns Hopkins): 5 195 CT volumes with nine abdominal organs annotated, per case a CT, a combined label map and one mask per organ. The repo is gated: accept its terms on huggingface.co and log in (HF_TOKEN); the package then fetches files on access. This manifest lists the first 100 cases explicitly (no repo listing needed); edit the sample list or use pl.load('hf://AbdomenAtlas/AbdomenAtlas1.0Mini/BDMAP_00001234') for others.

import pardesline as pl

ds = pl.load("hf_abdomenatlas_mini")
print(ds.describe())  # roles, frames, units, labels, licence, citation
s = ds[0]
obj = s.ct  # fetched now, cached under ~/.cache/pardesline/datasets/hf_abdomenatlas_mini

Provenance and licence

Source	Hugging Face Hub repo AbdomenAtlas/AbdomenAtlas1.0Mini, paths BDMAP_00000001
Scheme	hf
Size of the whole source	7.8 GB
Licence	CC-BY-NC-SA-4.0 (gated: accept the terms at the source first)
Commercial use	not allowed (non-commercial licence)
Redistribution	allowed
Licence note	CC BY-NC-SA 4.0, gated repository (checked 2026-10-14): non-commercial, share-alike, your own Hub token after accepting the terms.
Terms	https://huggingface.co/datasets/AbdomenAtlas/AbdomenAtlas1.0Mini
Homepage	https://github.com/MrGiovanni/AbdomenAtlas
Cite	Qu C., Zhang T., Qiao H., et al. AbdomenAtlas-8K: Annotating 8,000 CT Volumes for Multi-Organ Segmentation in Three Weeks. NeurIPS 2023. Li W., et al. AbdomenAtlas: A large-scale, detailed-annotated, & multi-center dataset for efficient transfer learning and open algorithmic benchmarking. Med Image Anal 97:103285, 2024.
Samples grouped by	id
Samples listed in the manifest	100

Roles (what each sample contains)

Role	Type	Frame	Units	Labels / layout	Pattern or file
ct	volume (CT)	from the file header	from the file header		BDMAP_00000001/ct.nii.gz, BDMAP_00000002/ct.nii.gz, BDMAP_00000003/ct.nii.gz Abdominal CT (NIfTI)
combined_labels	labels	from the file header	from the file header	1=aorta, 2=gall_bladder, 3=kidney_left, 4=kidney_right, 5=liver, 6=pancreas, 7=postcava, 8=spleen … (9 labels)	BDMAP_00000001/combined_labels.nii.gz, BDMAP_00000002/combined_labels.nii.gz, BDMAP_00000003/combined_labels.nii.gz Nine organs in one label map
aorta	labels	from the file header	from the file header	1=aorta	BDMAP_00000001/segmentations/aorta.nii.gz, BDMAP_00000002/segmentations/aorta.nii.gz, BDMAP_00000003/segmentations/aorta.nii.gz
gall_bladder	labels	from the file header	from the file header	1=gall_bladder	BDMAP_00000001/segmentations/gall_bladder.nii.gz, BDMAP_00000002/segmentations/gall_bladder.nii.gz, BDMAP_00000003/segmentations/gall_bladder.nii.gz
kidney_left	labels	from the file header	from the file header	1=kidney_left	BDMAP_00000001/segmentations/kidney_left.nii.gz, BDMAP_00000002/segmentations/kidney_left.nii.gz, BDMAP_00000003/segmentations/kidney_left.nii.gz
kidney_right	labels	from the file header	from the file header	1=kidney_right	BDMAP_00000001/segmentations/kidney_right.nii.gz, BDMAP_00000002/segmentations/kidney_right.nii.gz, BDMAP_00000003/segmentations/kidney_right.nii.gz
liver	labels	from the file header	from the file header	1=liver	BDMAP_00000001/segmentations/liver.nii.gz, BDMAP_00000002/segmentations/liver.nii.gz, BDMAP_00000003/segmentations/liver.nii.gz
pancreas	labels	from the file header	from the file header	1=pancreas	BDMAP_00000001/segmentations/pancreas.nii.gz, BDMAP_00000002/segmentations/pancreas.nii.gz, BDMAP_00000003/segmentations/pancreas.nii.gz
postcava	labels	from the file header	from the file header	1=postcava	BDMAP_00000001/segmentations/postcava.nii.gz, BDMAP_00000002/segmentations/postcava.nii.gz, BDMAP_00000003/segmentations/postcava.nii.gz
spleen	labels	from the file header	from the file header	1=spleen	BDMAP_00000001/segmentations/spleen.nii.gz, BDMAP_00000002/segmentations/spleen.nii.gz, BDMAP_00000003/segmentations/spleen.nii.gz
stomach	labels	from the file header	from the file header	1=stomach	BDMAP_00000001/segmentations/stomach.nii.gz, BDMAP_00000002/segmentations/stomach.nii.gz, BDMAP_00000003/segmentations/stomach.nii.gz

Standardized layout

Files are fetched on first access into the dataset cache and stay in their original format; ds.export(folder) writes every sample as standard files (PLY for points and meshes, MetaImage for volumes and label maps) with a .pardesline.json sidecar carrying frame, units, labels, licence and provenance, plus a manifest for the folder. Label maps become exact surfaces with to_mesh(label=k), DICOM series become one volume, raw arrays are decoded through the layout above.

ds = pl.load("hf_abdomenatlas_mini").take(2)
ds.export("./standardized")        # PLY / MetaImage + sidecars + manifest.json

Algorithms and models that run on it

• labels roles: mask.to_mesh(label=k), mask.to_numpy(), metrics: dice, hd95 (evaluation-metrics-3d) ; catalog modules: pyvista_volume_processing, simpleitk_registration, simpleitk_segmentation
• volume roles: volume.align(moving), volume.segment(lower, upper), volume.to_mesh(isovalue=...), volume.to_sitk() ; catalog modules: pyvista_volume_processing, simpleitk_registration, simpleitk_segmentation
• Supervised segmentation: prepare_nnunet_dataset.py writes this dataset in the nnU-Net v2 raw layout; PointNet trains on the label surfaces.
• Self-supervised pretraining: 3D JEPA reads the volumes directly.

Related

• Datasets catalog · How to download a medical imaging dataset in Python · Add a dataset
• Tags: medical, ct, abdomen, multi-organ, segmentation, nifti, hf, gated, non-commercial

_{Generated from src/pardesline/data/manifests/hf_abdomenatlas_mini.json on 2026-10-14.}