PAX-Ray++ dataset

folder cxas (4 files)
filepaxray_train_val.json.gz 264.97MB
filepaxray_test.json.gz 29.34MB
filepaxray_images_unfiltered.tar.gz 2.12GB 1.44GB
Type: Dataset

title= {PAX-Ray++ dataset},
keywords= {},
author= {},
abstract= {

Purpose: Interpreting chest radiographs (CXR) remains challenging due to the ambiguity of overlapping structures such as the lungs, heart, and bones. To address this issue, we propose a novel method for extracting fine-grained anatomical structures in CXR using pseudo-labeling of three-dimensional computed tomography (CT) scans.
Methods: We created a large-scale dataset of 10,021 thoracic CTs with 157 labels and applied an ensemble of 3D anatomy segmentation models to extract anatomical pseudo-labels. These labels were projected onto a two-dimensional plane, similar to the CXR, allowing the training of detailed semantic segmentation models for CXR without any manual annotation effort.
Results: Our resulting segmentation models demonstrated remarkable performance on CXR, with a high average model-annotator agreement between two radiologists with mIoU scores of 0.93 and 0.85 for frontal and lateral anatomy, while inter-annotator agreement remained at 0.95 and 0.83 mIoU. Our anatomical segmentations allowed for the accurate extraction of relevant explainable medical features such as the cardio-thoracic-ratio.
Conclusion: Our method of volumetric pseudo-labeling paired with CT projection offers a promising approach for detailed anatomical segmentation of CXR with a high agreement with human annotators. This technique may have important clinical implications, particularly in the analysis of various thoracic pathologies.

author    = {Constantin Seibold, Alexander Jaus, Matthias Fink,
Moon Kim, Simon ReiƟ, Jens Kleesiek*, Rainer Stiefelhagen*},
title     = {Accurate Fine-Grained Segmentation of Human Anatomy in Radiographs via Volumetric Pseudo-Labeling},
year      = {2023},
terms= {},
license= {},
superseded= {},
url= {}

Send Feedback