CheXpert Plus
Chest radiographs with full reports and 14 CheXpert pathology labels, including studies without those findings
Open the source page ↗ See it in the directory
At a glance
- What it is
- Chest radiographs with full reports and 14 CheXpert pathology labels, including studies without those findings
- Area
- Chest
- Modality
- DX, CR
- Format
- DICOM
- Size
- 223,462 radiograph–report pairs; 187,711 studies; 64,725 patients; about 6.8 TB
- Normals included
- Partial (mixed or unlabeled)
- License
- Stanford AIMI research use agreement (non-commercial; commercial license sold separately)
- Commercial use
- Not allowed under the free terms
- Access
- Free account and data-use terms required
- Status
- Verified: source page reviewed by Hounsfield Lab
- Added
- 2026-10-18
- Last checked
- 2026-11-11
- Source
- stanford.redivis.com/datasets/5yyj-1a9f6ap0x
How to access
Sign in on Redivis, open this dataset, and apply for access. Accept the Stanford research-use terms for this dataset (each dataset is a separate request). Non-commercial research is free. Commercial use needs a separate AIMI license: https://aimi.stanford.edu/datasets-commercial-use
Limitations
Free access is for non-commercial research under a data-use agreement. Do not redistribute the files or the download link. Do not attempt to re-identify anyone. A paid commercial license is a different contract; check the current fee and terms on the AIMI commercial-use page. DICOM plus the report text, split into sections. Labels include an uncertain class. This is the plus release (reports and DICOM), not the original CheXpert JPEG label release. A hospital mix, not healthy volunteers.
Citation required?
Yes. Cite the dataset DOI or the citation block on the Redivis page, and name Stanford AIMI.
Related open-source tools
- TorchXRayVision: Library of chest X-ray dataset loaders and pretrained classification, segmentation and autoencoder models
- CXAS: Chest X-ray anatomy segmentation (Mask2Former models: 187 frontal and 177 lateral classes) plus 132 named landmarks and landmark-based measurements such as cardiothoracic ratio, costophrenic angle, Cobb angle and the Haller index, in mm for DICOM input; cxas_idc writes one multi-segment DICOM SEG and a TID 1500 SR per radiograph; trained on PAXRay++ CT-projected radiographs
- CheXagent: Foundation model for chest X-ray interpretation
- MedRAX: Reasoning agent for chest X-rays that orchestrates multiple CXR models (ICML 2025)
- pydicom: Read, modify and write DICOM files in Python
- highdicom: High-level Python API on pydicom for reading images and writing AI results as DICOM Segmentation, Structured Report, parametric map and annotation objects
- OHIF Viewer: Zero-footprint web DICOM viewer
More chest datasets
- COCA: CT · Stanford AIMI research use agreement (non-commercial; commercial license sold separately)
- INSPECT: CT · Stanford AIMI research use agreement (non-commercial; commercial license sold separately)
- LDCT-and-Projection-data: CT · Chest, liver, and phantom images CC BY 4.0; head images under the NIH Controlled Data Access Policy
- LIDC-IDRI: CT, CR, DX · CC BY 3.0
- Medical Segmentation Decathlon: CT, MR · CC BY-SA 4.0
- MIDRC: CT, DX, CR, MR · University of Chicago data-use agreement (non-commercial research by default; a separate commercial agreement exists)
Hounsfield Lab links to the source and does not host images. The license and access summary reflects the source page on the last-checked date; terms can change, so read the source before you download. Not legal advice.