VinDr-RibCXR: A Benchmark Dataset for Automatic Segmentation and Labeling of Individual Ribs on Chest X-rays

TL;DR

This paper introduces VinDr-RibCXR, a new dataset with expert annotations for automatic rib segmentation and labeling on chest X-rays, providing a benchmark for future research in medical image analysis.

Contribution

The paper presents a new annotated dataset and baseline segmentation models for rib detection on chest X-rays, advancing research in automated medical imaging analysis.

Findings

Best model achieved a Dice score of 0.834.

The dataset contains 245 annotated chest X-rays.

Provides a baseline for future rib segmentation studies.

Abstract

We introduce a new benchmark dataset, namely VinDr-RibCXR, for automatic segmentation and labeling of individual ribs from chest X-ray (CXR) scans. The VinDr-RibCXR contains 245 CXRs with corresponding ground truth annotations provided by human experts. A set of state-of-the-art segmentation models are trained on 196 images from the VinDr-RibCXR to segment and label 20 individual ribs. Our best performing model obtains a Dice score of 0.834 (95% CI, 0.810--0.853) on an independent test set of 49 images. Our study, therefore, serves as a proof of concept and baseline performance for future research.