Explainable Artificial Intelligence for Breast Cancer Detection: A Review of Deep Learning, Multimodal Imaging, and Interpretable Diagnostic Frameworks

Breast cancer is a significant global health problem and early detection with accurate diagnosis is required. Computer-aided detection (CAD) systems based on Machine learning (ML) and Deep Learning (DL) have been developed. However, CAD systems suffer from poor interpretability which limits clinical translation. This manuscript provides an overview of Explainable Artificial Intelligence methods for breast cancer detection with medical imaging datasets and molecular marker datasets. Topics include multimodal deep learning, hybrid CNN-Transformer models, genetic algorithm design of interpretable ML models, and transfer learning coupled with explainability algorithms (Grad-CAM, LIME, Integrated Gradients, Occlusion Analysis). Classification accuracies of over 99% can be reached when using XAI-enabled deep learning models, without compromising transparency. Explainability techniques also allow clinicians to identify which regions of medical images are diagnostically relevant, improving trust and interpretability of AI models. Clinical decision support and precision oncology are potential application areas where XAI-integrated CAD systems can be utilized to their strengths. Integration of multimodal learning, federated learning, and image genomics represents a future direction for scalable, interpretable, and clinically deployable breast cancer detection solutions.