CheXpert Plus: Augmenting a Large Chest X-ray Dataset with Text Radiology Reports, Patient Demographics and Additional Image Formats

TL;DR

CheXpert Plus delivers the largest English-language radiology text-image resource to date by linking 223,228 chest X-ray images with 187,711 reports, 64,725 patients, and rich metadata including eight de-identified demographics, 14 pathology labels, and RadGraph annotations. The dataset provides images in DICOM and PNG, parsed report sections, and a longitudinal patient context, all de-identified and released under a research-use agreement, accompanied by pretrained models for RRG and RRS tasks. Core contributions include improved label extraction (CheXpert/CheXbert), RadGraph annotations, a robust de-identification workflow with extensive PHI auditing, and a Model Zoo featuring LLaMA, CLIP, VQ-GAN, and DINOv2 to support multimodal radiology research. Overall, CheXpert Plus enables cross-institution training, fairness-focused analyses with demographic data, and scalable evaluation for vision-language radiology tasks, potentially advancing diagnostic accuracy and patient care while acknowledging remaining biases and limitations.

Abstract

Since the release of the original CheXpert paper five years ago, CheXpert has become one of the most widely used and cited clinical AI datasets. The emergence of vision language models has sparked an increase in demands for sharing reports linked to CheXpert images, along with a growing interest among AI fairness researchers in obtaining demographic data. To address this, CheXpert Plus serves as a new collection of radiology data sources, made publicly available to enhance the scaling, performance, robustness, and fairness of models for all subsequent machine learning tasks in the field of radiology. CheXpert Plus is the largest text dataset publicly released in radiology, with a total of 36 million text tokens, including 13 million impression tokens. To the best of our knowledge, it represents the largest text de-identification effort in radiology, with almost 1 million PHI spans anonymized. It is only the second time that a large-scale English paired dataset has been released in radiology, thereby enabling, for the first time, cross-institution training at scale. All reports are paired with high-quality images in DICOM format, along with numerous image and patient metadata covering various clinical and socio-economic groups, as well as many pathology labels and RadGraph annotations. We hope this dataset will boost research for AI models that can further assist radiologists and help improve medical care. Data is available at the following URL: https://stanfordaimi.azurewebsites.net/datasets/5158c524-d3ab-4e02-96e9-6ee9efc110a1 Models are available at the following URL: https://github.com/Stanford-AIMI/chexpert-plus

Figures (2)

• Figure 1: Frequency plots for patient metadata as available in CheXpert Plus.
• Figure 2: The de-identification of images and reports from CheXpert Plus is an 8-step process as described in Section \ref{['sec:deidreport']}.