Machine Learning and Explainable AI for Breast Cancer Patient Prioritization: An Intelligent Decision-Support...

Breast cancer continues to represent a major global health burden, highlighting the need for effective approaches to risk stratification and clinical decision support. Conventional methods, including the Breast Imaging Reporting and Data System (BI-RADS) and histopathological classifications, primarily rely on clinical assessments and may not fully account for relevant demographic and behavioral characteristics. To overcome these limitations, we present an integrated framework combining K-Means clustering, Random Forest classification, and Explainable Artificial Intelligence (XAI) to support breast cancer risk stratification and patient prioritization. The proposed methodology uses clustering to stratify patients into low-, medium-, and high-risk groups, followed by supervised machine learning to reproduce the cluster-derived risk categories, achieving an accuracy of 98%. To enhance interpretability, Local Interpretable Model-Agnostic Explanations (LIME) are incorporated to identify the variables that most strongly influence individual classifications, including body mass index (BMI), breastfeeding practices, and maternal age. By integrating multiple dimensions of patient information, the framework provides a more comprehensive characterization of risk while increasing the transparency of the decision-making process. Its relatively simple and scalable architecture also facilitates potential implementation in healthcare environments with limited resources. Simulation experiments further provide a proof-of-concept evaluation of the proposed prioritization approach. Compared with random patient selection, the strategy achieved a substantially higher average severity score (1.66 vs. 0.92) and prioritized 4.3 times more high-risk patients. These findings suggest that the proposed framework can serve as an intelligent decision-support tool for prioritizing breast cancer patients and improving resource allocation when healthcare capacity is constrained.