Towards Early Detection: AI-Based Five-Year Forecasting of Breast Cancer Risk Using Digital Breast Tomosynthesis Imaging

2025-08-31

AI Summary

Current breast cancer risk prediction models exhibit limited performance and fail to leverage digital breast tomosynthesis (DBT), an emerging imaging modality. To address this, we propose the first DBT-based five-year breast cancer risk prediction framework. Our method introduces Meta AI’s DINOv2 self-supervised image encoder—deployed for the first time in DBT feature extraction—integrated with a deep learning backbone and a differentiable cumulative hazard layer to enable end-to-end personalized risk modeling. Trained on a large-scale real-world DBT cohort, the model achieves 0.80 AUROC on an independent test set, significantly outperforming existing clinical and radiological benchmarks. This work demonstrates DBT’s substantial information gain for long-term risk prediction and establishes a novel paradigm integrating self-supervised representation learning with survival analysis. By bridging advanced computer vision and clinical survival modeling, our framework provides a clinically actionable pathway toward precision breast cancer screening.

Technology Category

• Computer Vision: Representation Learning for Vision
• Machine Learning: Calibration & Uncertainty Quantification
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

As early detection of breast cancer strongly favors successful therapeutic outcomes, there is major commercial interest in optimizing breast cancer screening. However, current risk prediction models achieve modest performance and do not incorporate digital breast tomosynthesis (DBT) imaging, which was FDA-approved for breast cancer screening in 2011. To address this unmet need, we present a deep learning (DL)-based framework capable of forecasting an individual patient's 5-year breast cancer risk directly from screening DBT. Using an unparalleled dataset of 161,753 DBT examinations from 50,590 patients, we trained a risk predictor based on features extracted using the Meta AI DINOv2 image encoder, combined with a cumulative hazard layer, to assess a patient's likelihood of developing breast cancer over five years. On a held-out test set, our best-performing model achieved an AUROC of 0.80 on predictions within 5 years. These findings reveal the high potential of DBT-based DL approaches to complement traditional risk assessment tools, and serve as a promising basis for additional investigation to validate and enhance our work.

Problem

• Predicting 5-year breast cancer risk using AI
• Incorporating digital breast tomosynthesis imaging data
• Improving current limited performance risk models

Innovation

• Deep learning framework for breast cancer risk
• Uses DINOv2 image encoder on DBT scans
• Predicts 5-year risk with cumulative hazard layer