LNDb19

Curated release of LNDb (Lung Nodule Database) public chest CT data for research use.

Dataset viewer (Hugging Face)

The Dataset viewer loads data/train-00000-of-00001.parquet with subject_id, ct_path, nodule_mask_paths (semicolon-separated), and num_nodule_masks. Values are strings/paths only so the Hub skips NIfTI embedding that previously failed. Download .nii.gz from the Files tab.

OpenVAE / SMILE project

This dataset is published alongside the OpenVAE open-source stack for medical CT/MRI latent modeling and diffusion-style pipelines. The same public cohorts are used in our benchmarking and demo workflows.

• OpenVAE (code): github.com/KumaKuma2002/OpenVAE
• OpenVAE (pretrained VAEs on Hugging Face): huggingface.co/SMILE-project/OpenVAE

Citation

If you use this resource together with OpenVAE, SMILE-style preprocessing, or our benchmarks, please cite:

@article{liu2025see,
  title={See More, Change Less: Anatomy-Aware Diffusion for Contrast Enhancement},
  author={Liu, Junqi and Wu, Zejun and Bassi, Pedro RAS and Zhou, Xinze and Li, Wenxuan and Hamamci, Ibrahim E and Er, Sezgin and Lin, Tianyu and Luo, Yi and P{\l}otka, Szymon and others},
  journal={arXiv preprint arXiv:2512.07251},
  year={2025}
}

Paper: arxiv.org/abs/2512.07251

Please also cite the original LNDb / LUNA19 challenge materials and follow the organizers’ license and attribution requirements.