NIH ChestX-ray14

Unknown

112,120 frontal-view chest X-ray images from 30,805 unique patients with 14 disease labels extracted using NLP from radiology reports. Foundational benchmark for chest X-ray AI.

Benchmark Stats

Models4
Papers4
Metrics1

SOTA History

Not enough data to show trend.

Only 4 models on this benchmark

Help build the community leaderboard — submit your model results.

auroc

auroc

Higher is better

RankModelSourceScoreYearPaper
1torchxrayvision

Multi-dataset pre-training improves over single-dataset.

Editorial85.82025Source
2chexnet

Original CheXNet on ChestX-ray14. Exceeded radiologist performance on pneumonia (0.768 vs 0.633).

Editorial84.12025Source
3densenet-121-cxr

Original NIH baseline model.

Editorial82.62025Source
4resnet-50-cxr

ResNet-50 baseline.

Editorial80.42025Source

Submit a Result