Radio-opaque artefacts in digital mammography: automatic detection and analysis of downstream effects

TL;DR

This paper presents an automatic detection system for radio-opaque artefacts in digital mammography and demonstrates their significant impact on classification model performance, emphasizing the need for robust artefact handling.

Contribution

It introduces a multi-label artefact detector for mammograms and analyzes how artefacts affect model accuracy and thresholds, advancing reliability in digital mammography analysis.

Findings

Artefacts significantly impact model performance.

Automatic detection improves robustness.

Publicly available annotations and models facilitate future research.

Abstract

This study investigates the effects of radio-opaque artefacts, such as skin markers, breast implants, and pacemakers, on mammography classification models. After manually annotating 22,012 mammograms from the publicly available EMBED dataset, a robust multi-label artefact detector was developed to identify five distinct artefact types (circular and triangular skin markers, breast implants, support devices and spot compression structures). Subsequent experiments on two clinically relevant tasks $-$ breast density assessment and cancer screening $-$ revealed that these artefacts can significantly affect model performance, alter classification thresholds, and distort output distributions. These findings underscore the importance of accurate automatic artefact detection for developing reliable and robust classification models in digital mammography. To facilitate future research our annotations, code, and model predictions are made publicly available.