Exploring the Role of XAI in Enhancing Predictive Model Transparency in Healthcare Risk Assessment
Ramya Mandava, Sai Srinivas Vellela, N. Malathi, Koya Haritha, Shobana Gorintla, Lavanya Dalavai
Abstract
The increasing integration of Artificial Intelligence (AI) in healthcare demands systems that not only predict accurately but also explain their decisions transparently. This study explores the role of Explainable Artificial Intelligence (XAI) in enhancing transparency, interpretability, and trust in AI-driven predictive models used for healthcare risk assessment. It evaluates XAI techniques like SHAP and LIME in explaining black-box models such as Random Forest and Neural Networks, focusing on clinician trust, ethical adoption, and patient safety. Utilizing real-world healthcare datasets, the study investigates how interpretable models can bridge the gap between AI outputs and clinical decision-making. Results demonstrate that XAI significantly improves clinician confidence and regulatory compliance without greatly compromising prediction performance. The research underscores the transformative potential of XAI in fostering responsible and effective AI deployment in clinical environments.