Meta-repository of screening mammography classifiers

TL;DR

This paper introduces a meta-repository for screening mammography classifiers that facilitates reproducibility, comparison, and evaluation of AI models across diverse datasets, supporting progress in breast cancer screening research.

Contribution

It provides a unified framework with open-source models and evaluation tools, enabling standardized comparison and reproducibility in mammography AI research.

Findings

Performance comparison of five models on seven datasets

Framework's flexibility for other medical imaging tasks

Open-source implementation promotes reproducibility

Abstract

Artificial intelligence (AI) is showing promise in improving clinical diagnosis. In breast cancer screening, recent studies show that AI has the potential to improve early cancer diagnosis and reduce unnecessary workup. As the number of proposed models and their complexity grows, it is becoming increasingly difficult to re-implement them. To enable reproducibility of research and to enable comparison between different methods, we release a meta-repository containing models for classification of screening mammograms. This meta-repository creates a framework that enables the evaluation of AI models on any screening mammography data set. At its inception, our meta-repository contains five state-of-the-art models with open-source implementations and cross-platform compatibility. We compare their performance on seven international data sets. Our framework has a flexible design that can be generalized to other medical image analysis tasks. The meta-repository is available at https://www.github.com/nyukat/mammography_metarepository.