ChestX-ray14

Dataset Information
Modalities
Images, Medical
Introduced
2017
License
Unknown
Homepage

Overview

ChestX-ray14 is a medical imaging dataset which comprises 112,120 frontal-view X-ray images of 30,805 (collected from the year of 1992 to 2015) unique patients with the text-mined fourteen common disease labels, mined from the text radiological reports via NLP techniques. It expands on ChestX-ray8 by adding six additional thorax diseases: Edema, Emphysema, Fibrosis, Pleural Thickening and Hernia.

Source: https://nihcc.app.box.com/v/ChestXray-NIHCC/file/220660789610
Image Source: https://nihcc.app.box.com/v/ChestXray-NIHCC

Variants: ChestX-ray14, ChestXray14 1024x1024

Associated Benchmarks

This dataset is used in 4 benchmarks:

Recent Benchmark Submissions

Task Model Paper Date
Multi-Label Classification Improved CheXNet (DannyNet, dstrick17 et al., 2025) Reproducing and Improving CheXNet: Deep … 2025-05-10
Multi-Task Learning BayesAgg-MTL Bayesian Uncertainty for Gradient Aggregation … 2024-02-06
Medical Image Generation StyleGAN2 with DiffAugment Feature Extraction for Generative Medical … 2023-11-22
Multi-Label Classification SynthEnsemble SynthEnsemble: A Fusion of CNN, … 2023-11-13
Multi-Label Classification CoAtNet SynthEnsemble: A Fusion of CNN, … 2023-11-13
Pneumonia Detection MUXNet-m MUXConv: Information Multiplexing in Convolutional … 2020-03-31
Multi-Label Classification DensNet121 CheXclusion: Fairness gaps in deep … 2020-02-14
Pneumonia Detection NSGANetV1-X Multi-Objective Evolutionary Design of Deep … 2019-12-03
Pneumonia Detection NSGANetV1-A3 Multi-Objective Evolutionary Design of Deep … 2019-12-03
Pneumonia Detection CheXNet CheXNet: Radiologist-Level Pneumonia Detection on … 2017-11-14

Research Papers

Recent papers with results on this dataset: