TorchXRayVision: A library of chest X-ray datasets and models

TL;DR

TorchXRayVision is an open source library that standardizes access to diverse chest X-ray datasets and models, facilitating research and development in medical imaging analysis.

Contribution

It introduces a unified interface and pre-processing pipeline for multiple chest X-ray datasets and provides various pre-trained models for classification and feature extraction.

Findings

Streamlines access to multiple datasets and models

Enables consistent pre-processing across datasets

Supports baseline and feature extraction tasks

Abstract

TorchXRayVision is an open source software library for working with chest X-ray datasets and deep learning models. It provides a common interface and common pre-processing chain for a wide set of publicly available chest X-ray datasets. In addition, a number of classification and representation learning models with different architectures, trained on different data combinations, are available through the library to serve as baselines or feature extractors.