# Enhancing Breast Cancer Risk Prediction by Incorporating Prior Images

arXiv: https://arxiv.org/abs/2303.15699 · HTML: https://mlsanity.com/paper/2303.15699/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 32 references that are in the map.
- First posted on arXiv 28 Mar 2023.

## Builds on (most cited first, 7 of 7)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [SGDR: Stochastic Gradient Descent with Warm Restarts](https://mlsanity.com/paper/1608.03983/) (2016)
- [DeepSurv: Personalized Treatment Recommender System Using A Cox Proportional Hazards Deep Neural Network](https://mlsanity.com/paper/1606.00931/) (2016)
- [Decoupling Inherent Risk and Early Cancer Signs in Image-based Breast Cancer Risk Models](https://mlsanity.com/paper/2007.05791/) (2020)
- [Risk models for breast cancer and their validation](https://mlsanity.com/paper/1907.02829/) (2019)
- [Screening Mammogram Classification with Prior Exams](https://mlsanity.com/paper/1907.13057/) (2019)

## Built on by (most cited first, 15 of 15)

- [Longitudinal Mammogram Risk Prediction](https://mlsanity.com/paper/2404.19083/) (2024)
- [Ordinal Learning: Longitudinal Attention Alignment Model for Predicting Time to Future Breast Cancer Events from Mammograms](https://mlsanity.com/paper/2409.06887/) (2024)
- [Reconsidering Explicit Longitudinal Mammography Alignment for Enhanced Breast Cancer Risk Prediction](https://mlsanity.com/paper/2506.19363/) (2025)
- [VMRA-MaR: An Asymmetry-Aware Temporal Framework for Longitudinal Breast Cancer Risk Prediction](https://mlsanity.com/paper/2506.17412/) (2025)
- [Assessing the significance of longitudinal data in Alzheimer's Disease forecasting](https://mlsanity.com/paper/2405.17352/) (2024)
- [STA-Risk: A Deep Dive of Spatio-Temporal Asymmetries for Breast Cancer Risk Prediction](https://mlsanity.com/paper/2505.21699/) (2025)
- [The Impact of Longitudinal Mammogram Alignment on Breast Cancer Risk Assessment](https://mlsanity.com/paper/2511.08328/) (2025)
- [Longitudinal Multi-View Breast Cancer Risk Prediction](https://mlsanity.com/paper/2607.11343/) (2026)
- [Longitudinal Mammogram Exam-based Breast Cancer Diagnosis Models: Vulnerability to Adversarial Attacks](https://mlsanity.com/paper/2411.00837/) (2024)
- [Towards Early Detection: AI-Based Five-Year Forecasting of Breast Cancer Risk Using Digital Breast Tomosynthesis Imaging](https://mlsanity.com/paper/2509.00900/) (2025)
- [The LongiMam model for improved breast cancer risk prediction using longitudinal mammograms](https://mlsanity.com/paper/2509.21383/) (2025)
- [Breast Cancer VLMs: Clinically Practical Vision-Language Train-Inference Models](https://mlsanity.com/paper/2510.25051/) (2025)
- [MV-MLM: Bridging Multi-View Mammography and Language for Breast Cancer Diagnosis and Risk Prediction](https://mlsanity.com/paper/2510.26151/) (2025)
- [MGRegBench: A Novel Benchmark Dataset with Anatomical Landmarks for Mammography Image Registration](https://mlsanity.com/paper/2512.17605/) (2025)
- [Explicitly Aligned Multi-View Breast Cancer Risk Prediction from Variable-Length Screening Histories](https://mlsanity.com/paper/2610.16329/) (2026)

## Abstract (arXiv)

Recently, deep learning models have shown the potential to predict breast cancer risk and enable targeted screening strategies, but current models do not consider the change in the breast over time. In this paper, we present a new method, PRIME+, for breast cancer risk prediction that leverages prior mammograms using a transformer decoder, outperforming a state-of-the-art risk prediction method that only uses mammograms from a single time point. We validate our approach on a dataset with 16,113 exams and further demonstrate that it effectively captures patterns of changes from prior mammograms, such as changes in breast density, resulting in improved short-term and long-term breast cancer risk prediction. Experimental results show that our model achieves a statistically significant improvement in performance over the state-of-the-art based model, with a C-index increase from 0.68 to 0.73 (p < 0.05) on held-out test sets.