Artificial Intelligence in Modern Diagnostic Systems: From Explainable Models to Clinical and Industrial Applications

Abstract

Artificial Intelligence (AI) has emerged as the central engine of next-generation diagnostic systems, reshaping medical imaging, precision agriculture, and intelligent industrial maintenance. This review presents a unified technical perspective on computer-aided diagnosis (CAD) empowered by deep learning and Explainable AI (XAI). We synthesize methodologies ranging from convolutional neural networks (CNN), transfer learning, hybrid multimodal fusion, to generative language assistants. Mathematical foundations of learning, optimization, and attribution are detailed alongside pseudocode for reproducible pipelines. Comparative tables analyze algorithms across accuracy, interpretability, computational cost, and regulatory readiness.