A Comprehensive Literature Survey on CNN- Based Breast Cancer Prognosis Using Multi-Modal Data

Breast cancer remains one of the most prevalent diseases affecting women worldwide. Accurate prognosis plays a vital role in guiding treatment decisions and improving survival rates. In recent years, Convolutional Neural Networks (CNNs) have gained significant attention for their ability to automate diagnostic and prognostic tasks. This paper reviews recent CNN-based models developed for breast cancer prognosis, particularly those integrating multi-modal data such as clinical, imaging, and molecular profiles. We explore key trends in model design, data fusion strategies, and common datasets used in research. Although CNNs show promising results, challenges such as limited interpretability and poor generalization remain. To address these, we suggest future research directions involving attention-based data fusion and explainable CNN architectures, with the goal of enhancing clinical adoption and reliability.