Abstract

Breast cancer is a major health concern affecting women worldwide. Early detection and accurate prediction of breast cancer risk are crucial for improving patient outcomes. This study focuses on leveraging deep learning techniques to analyze longitudinal mammography examinations for predicting breast cancer risk. The proposed method utilizes a large dataset of mammograms from multiple time points for each patient, allowing for the extraction of temporal patterns and trends in breast tissue changes. By training a deep learning model on this longitudinal data, we aim to develop a predictive model capable of identifying individuals at higher risk of developing breast cancer. The model is evaluated on an independent dataset, and its performance is compared with traditional risk assessment methods. The results demonstrate the potential of deep learning in leveraging temporal information from longitudinal mammography examinations to accurately predict breast cancer risk. This approach has the potential to enhance existing risk assessment models and facilitate personalized screening and prevention strategies.