Longitudinal Multi-View Breast Cancer Risk Prediction

2026-07-13

AI Summary

Current approaches to breast cancer risk prediction struggle to effectively integrate spatial and temporal information from the two anatomically complementary mammographic views—craniocaudal (CC) and mediolateral oblique (MLO)—along with longitudinal imaging sequences. This work proposes LMV-Net, a novel framework that, for the first time, combines an explicit longitudinal alignment mechanism with multi-view modeling to jointly analyze complementary features across both views and time within a unified architecture. By leveraging a deep network to achieve temporal alignment and multi-view fusion, LMV-Net significantly enhances risk stratification performance. Evaluated on the EMBED and CSAW-CC datasets, the model consistently outperforms state-of-the-art methods and demonstrates robust advantages across diverse breast density categories and cancer subtypes.

Abstract

Accurate breast cancer risk prediction from screening mammography is critical for enabling personalized screening intervals and early detection. Recent deep learning methods have shown the value of longitudinal data and explicit temporal alignment. However, existing approaches either perform explicit alignment using a single mammographic view or model multiple views without explicit longitudinal alignment, limiting their ability to exploit the complementary spatial-temporal information used in clinical practice. To address this gap, we propose LMV-Net, a longitudinal multi-view breast cancer risk prediction model that jointly analyzes anatomically complementary CC and MLO views within an explicitly aligned longitudinal framework. We evaluate our approach on the public EMBED and CSAW-CC datasets, comparing it to state-of-the-art breast cancer risk prediction methods. Our model consistently outperforms existing approaches in overall risk prediction performance and across different breast density and cancer subgroups. Importantly, these improvements highlight the potential of longitudinal multi-view modeling to enhance risk stratification, paving the way for future work on personalized screening, earlier identification of high-risk patients, and more efficient screening resource allocation. The code is available at https://github.com/sot176/LMV-Net.

Problem

• breast cancer risk prediction
• longitudinal data
• multi-view mammography
• temporal alignment
• spatial-temporal information

Innovation

• longitudinal alignment
• multi-view mammography
• breast cancer risk prediction
• deep learning
• temporal modeling