Evolution of Explainable AI in Healthcare: Toward Trustworthy and Accurate Diagnostics

Abstract

Healthcare diagnostics, treatment personalization, and clinical decision-making are being transformed by AI. This work summarizes recent literature to assess the current state of AI in healthcare. Recent AI-driven healthcare applications are examined with respect to diagnostic accuracy and algorithmic transparency. Our analysis of recent research indicates that ensemble machine learning models and deep learning approaches achieve high accuracy, ranging from 83% to over 99%, across several medical domains. Explainable AI methods, such as SHAP and LIME, are used to address the black-box challenge, thereby enhancing clinical trust and acceptance. Emerging issues highlighted in this review include data privacy, algorithmic bias and regulatory frameworks. Promising research directions that balance technological innovation with ethical responsibility are suggested.