Enhancing Transparency in Healthcare Analytics Through Explainable Association Rule Mining

Abstract

The rapid expansion of medical data from electronic health records, imaging, and genomics has amplified the need for interpretable artificial intelligence in healthcare. This chapter presents the integration of Explainable AI (XAI) with Association Rule Mining (ARM) to enhance transparency, trust, and ethical accountability in clinical decision-making. It explores classical ARM techniques and their transformation through XAI frameworks such as SHAP, LIME, Anchors, and RuleFit, enabling clinicians to understand not only what the model predicts but also why. Case studies on disease prediction, drug interaction detection, and clinical decision support demonstrate how XAI-ARM bridges data-driven insights with medical reasoning. Furthermore, the chapter discusses evaluation metrics for explainability and outlines ethical, legal, and practical frameworks for deploying trustworthy, human-centered AI in healthcare systems.

Ähnliche Arbeiten