Explainable Artificial Intelligence in Medical Diagnosis: Leveraging Data Mining for Transparent Disease Prediction

Abstract

The integration of Artificial Intelligence (AI) in healthcare has significantly enhanced the efficiency and accuracy of medical diagnosis. However, the black-box nature of most AI models raises serious concerns regarding trust, transparency, and accountability—especially in high-stakes domains like healthcare. This paper explores the role of Explainable AI (XAI) techniques, particularly SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-Agnostic Explanations), in making disease prediction models more interpretable for clinicians and stakeholders. It also examines the application of data mining techniques to extract meaningful patterns from medical datasets, which serve as a foundation for building accurate and explainable diagnostic tools. By integrating machine learning with explainable frameworks, this study aims to bridge the gap between model accuracy and interpretability, thereby fostering trust in AI-assisted decision-making. Real world case studies, quantitative evaluations, and model interpretation visualizations are presented to demonstrate the practical impact of XAI in enhancing diagnostic transparency.

Keywords: Explainable AI, healthcare, data mining, diagnosis, XAI, machine learning