Skip to content

LUNA16: 888 thoracic CTs (LIDC-IDRI subset) for lung nodule detection

pl.load("luna16") loads this dataset. The LUNA16 challenge scans: 888 CTs from LIDC-IDRI with slice thickness <= 2.5 mm, as MetaImage (.mhd + .raw), in ten subsets for cross-validation. Each subset is a 6-7 GB zip on Zenodo read scan by scan with HTTP Range (the .raw sibling is fetched with its header). Nodule annotations (annotations.csv, candidates.csv) are separate files on the Zenodo record.

import pardesline as pl

ds = pl.load("luna16")
print(ds.describe())  # roles, frames, units, labels, licence, citation
s = ds[0]
obj = s.ct  # fetched now, cached under ~/.cache/pardesline/datasets/luna16

Provenance and licence
Source remote archive read by HTTP Range: https://zenodo.org/records/3723295/files/subset0.zip, https://zenodo.org/records/3723295/files/subset1.zip, https://zenodo.org/records/3723295/files/subset2.zip, https://zenodo.org/records/3723295/files/subset3.zip, https://zenodo.org/records/3723295/files/subset4.zip, https://zenodo.org/records/3723295/files/subset5.zip, https://zenodo.org/records/3723295/files/subset6.zip, https://zenodo.org/records/4121926/files/subset7.zip, https://zenodo.org/records/4121926/files/subset8.zip, https://zenodo.org/records/4121926/files/subset9.zip
Scheme archive
Size of the whole source 64.8 GB
Licence CC-BY-4.0
Commercial use allowed by the licence
Redistribution allowed
Licence note CC BY 4.0 on the Zenodo records (checked 2026-10-14); the underlying LIDC-IDRI collection is CC BY 3.0 (TCIA).
Terms https://zenodo.org/records/3723295
Homepage https://luna16.grand-challenge.org/
Cite Setio A.A.A., Traverso A., de Bel T., et al. Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: the LUNA16 challenge. Med Image Anal 42:1-13, 2017.
Samples grouped by subset

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 subset{subset}/{id}.mhd
Thorax CT (MetaImage header + raw; LPS mm from the header); field subset = fold 0..9

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("luna16").take(2)
ds.export("./standardized")        # PLY / MetaImage + sidecars + manifest.json

Algorithms and models that run on it

Generated from src/pardesline/data/manifests/luna16.json on 2026-10-14.