BPD-Neo: neonatal lung MRI with lung and trachea masks for bronchopulmonary dysplasia (BPD), 40 infants

Provider
Rachit Saluja et al. (Weill Cornell Medicine / Cornell); mirrored by MedOtter
License
CC-BY-4.0 (source)
Size
83 files
Format
Parquet, NIfTI (in sampled files)
Stored at
Hugging Face (https://huggingface.co/datasets/MedOtter/BPD-Neo)
Access
Free and public. No account with the data provider is needed.
DOI
10.5281/zenodo.15768091
Category
Health & Biomedical
Source
https://doi.org/10.1038/s41597-026-07006-8

About this dataset

BPD-Neo is a neonatal lung MRI dataset from Weill Cornell Medicine and Cornell of 40 newborns, most diagnosed with bronchopulmonary dysplasia (BPD), a chronic lung disease of preterm infants that is usually judged on portable chest X-rays in the NICU. Each case has an MRI volume with a label map of lung and trachea (labels 1 and 2, 0 background), and the release ships clinical data and baseline segmentation models that the authors checked against clinical assessments. Volumes and masks are NIfTI under train/ with previews in preview/; about 80 MB (measured 2026-11-21). Read with nibabel or SimpleITK. Use for neonatal lung and airway segmentation, quantitative lung volume measurement in preterm infants and radiation-free BPD assessment research. Code and pretrained weights are at github.com/rachitsaluja/BPD-Neo. Mirrored by MedOtter from the authors' Zenodo record (v1, June 2025), CC BY 4.0, DOI 10.5281/zenodo.15768091; cite Saluja et al., Scientific Data 2026 (doi 10.1038/s41597-026-07006-8).

How to load this dataset in Python

from datasets import load_dataset
ds = load_dataset("MedOtter/BPD-Neo")
print(ds)

Data structure

File types in sampled files: Parquet, NIfTI

Files

Example file: preview/train-00000-of-00001.parquet

Analyze this dataset without downloading it

On Scigantic, BPD-Neo: neonatal lung MRI with lung and trachea masks for bronchopulmonary dysplasia (BPD), 40 infants opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.

This page is also available as Markdown and as JSON.

Related Health & Biomedical datasets