MammoDG: Generalisable Deep Learning Breaks the Limits of Cross-Domain Multi-Center Breast Cancer Screening
This work addresses the need for reliable and generalizable AI tools to support radiologists in breast cancer screening, reducing workload and improving diagnostic accuracy across diverse clinical settings, though it is incremental in improving domain generalization for a specific medical imaging task.
The paper tackles the problem of high variability in mammograms across different imaging protocols and centers, which reduces the sensitivity of existing machine learning models, by introducing MammoDG, a deep-learning framework that uses multi-view mammograms and a novel contrastive mechanism to enhance generalization, achieving superior performance in cross-domain multi-center breast cancer screening.
Breast cancer is a major cause of cancer death among women, emphasising the importance of early detection for improved treatment outcomes and quality of life. Mammography, the primary diagnostic imaging test, poses challenges due to the high variability and patterns in mammograms. Double reading of mammograms is recommended in many screening programs to improve diagnostic accuracy but increases radiologists' workload. Researchers explore Machine Learning models to support expert decision-making. Stand-alone models have shown comparable or superior performance to radiologists, but some studies note decreased sensitivity with multiple datasets, indicating the need for high generalisation and robustness models. This work devises MammoDG, a novel deep-learning framework for generalisable and reliable analysis of cross-domain multi-center mammography data. MammoDG leverages multi-view mammograms and a novel contrastive mechanism to enhance generalisation capabilities. Extensive validation demonstrates MammoDG's superiority, highlighting the critical importance of domain generalisation for trustworthy mammography analysis in imaging protocol variations.