Skip to content

TCIA Pancreas-CT: 82 abdominal contrast-enhanced CTs (NIH)

pl.load("tcia_pancreas_ct") loads this dataset. 82 portal-venous CT scans of the pancreas from the NIH Clinical Center, one DICOM series per patient (the pancreas label maps of the original release are NIfTI files on the collection page, not in NBIA). Series are listed through the public NBIA API (no account) and each one is downloaded as a zip of DICOM files on first access; the folder is then a DICOM volume (LPS mm from the headers).

import pardesline as pl

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

Provenance and licence
Source The Cancer Imaging Archive, collection Pancreas-CT through the NBIA REST API
Scheme tcia
Size of the whole source 8.8 GB
Licence CC-BY-3.0
Commercial use allowed by the licence
Redistribution allowed
Licence note CC BY 3.0 on the TCIA collection page (checked 2026-10-14).
Terms https://www.cancerimagingarchive.net/collection/pancreas-ct/
Homepage https://www.cancerimagingarchive.net/collection/pancreas-ct/
Cite Roth H.R., Farag A., Turkbey E.B., et al. Data From Pancreas-CT. The Cancer Imaging Archive, 2016. doi:10.7937/K9/TCIA.2016.tNB1kqBU. Clark K., Vendt B., Smith K., et al. The Cancer Imaging Archive (TCIA): Maintaining and Operating a Public Information Repository. J Digit Imaging 26(6):1045-1057, 2013.
Samples grouped by id

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 {id}/CT/{series}
Abdominal CT, DICOM series

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("tcia_pancreas_ct").take(2)
ds.export("./standardized")        # PLY / MetaImage + sidecars + manifest.json

Algorithms and models that run on it

Generated from src/pardesline/data/manifests/tcia_pancreas_ct.json on 2026-10-14.