Resources

CXLSeg dataset: Chest X-ray with Lung Segmentation, a comparatively large dataset of segmented Chest X-ray radiographs based on the MIMIC-CXR dataset. This contains segmentation results of 243,324 frontal view images and corresponding masks.

CheXStruct is an automated pipeline that derives structured diagnostic reasoning steps from chest X-rays. CXReasonBench builds on this to evaluate whether models perform clinically grounded, multi-step reasoning beyond final diagnoses.

CXR-LT 2024 was a challenge for long-tailed, multi-label, and zero-shot thorax disease classification on chest X-rays, held at MICCAI 2024. This page contains long-tailed labels for 45 diseases from the CXR-LT 2024 and 2023 challenges.

BRAX contains 24,959 chest radiography exams and 40,967 images acquired in a large general Brazilian hospital. All images have been read by trained radiologists and 14 labels were derived from Brazilian Portuguese reports using NLP.

This dataset contains 3032 cases of eye-tracking data collected while five radiologists dictated reports for frontal chest x-rays, synchronized timestamped dictation transcription, and manual labels for validation of localization of abnormalities.

CXR-LT 2024 was a challenge for long-tailed, multi-label, and zero-shot thorax disease classification on chest X-rays, held at MICCAI 2024. This page contains long-tailed labels for 45 diseases from the CXR-LT 2024 and 2023 challenges.

This dataset includes bounding box-statement pairs for chest X-ray images, derived from radiologists’ eye-tracking data (for explainability) and annotations, for local visual-language models.

This dataset was a collected using an eye tracking system while a radiologist interpreted and read 1,083 public CXR images. The dataset contains the following aligned modalities: image, transcribed report text, dictation audio and eye gaze data.

CheXStruct is an automated pipeline that derives structured diagnostic reasoning steps from chest X-rays. CXReasonBench builds on this to evaluate whether models perform clinically grounded, multi-step reasoning beyond final diagnoses.

This dataset includes bounding box-statement pairs for chest X-ray images, derived from radiologists’ eye-tracking data (for explainability) and annotations, for local visual-language models.