Artificial Intelligence–Enabled Wearable Sensors for Continuous Health Monitoring-An Updated Review for Biomedical Engineering

Abstract

Background: Wearable biosensors integrated with artificial intelligence (AI) have significantly advanced continuous health monitoring by enabling real-time, personalized, and non-invasive assessment of physiological and behavioral parameters beyond traditional clinical environments. These technologies support disease management, early diagnosis, preventive care, and personalized interventions across diverse health domains. Aim: This review aims to summarize recent advancements in AI-enabled wearable biosensors, focusing on their applications, methodological innovations, challenges, and future directions in biomedical engineering. Methods: A comprehensive narrative review of recent scientific literature was conducted, analyzing developments in wearable sensor technologies, AI methodologies (including machine learning, deep learning, edge AI, federated learning, and human-in-the-loop systems), and their applications across metabolic, cardiovascular, neurological, and neonatal health domains. Results: AI-powered wearable biosensors demonstrate high potential for continuous health monitoring, predictive analytics, and personalized intervention. Applications include glucose monitoring, cardiovascular risk detection, gait and motor assessment, and neonatal surveillance. Advances in edge computing and federated learning enhance privacy and real-time responsiveness, while digital twins and large language models improve interpretability and decision support. Conclusion: AI-enabled wearable biosensors are transforming healthcare toward predictive, proactive, and personalized models of care, although challenges related to data privacy, robustness, biological integration, and regulation remain.