Artificial Intelligence Empowers Monitoring, Assessment, and Management in Exercise

The rapid development of artificial intelligence (AI) is a powerful catalyst for personalized exercise monitoring and management. This narrative review examines how exercise and physiological indicators are used in health surveillance and disease management and evaluates the integration of AI in these domains. Accumulating evidence demonstrates strong associations between indicators such as gait, electromyography, and cardiorespiratory fitness with disease outcomes and overall health. AI methods are increasingly being applied to improve monitoring, diagnosis, and personalized exercise prescription, offering greater precision compared with traditional approaches. Nevertheless, challenges persist, including limited data quality, lack of multimodal integration, concerns with interpretability, and ethical issues surrounding privacy. Many studies still rely on small cohorts, limiting clinical applicability. We synthesize current findings, identify research gaps, and propose future directions to enhance the robustness and practical value of AI-driven approaches in exercise and health research.