Abstract

This article provides a comprehensive review of IoT-based smart health risk prediction systems, which integrate Artificial Intelligence (AI) and biomedical sensors to revolutionize healthcare delivery, specifically for chronic disease management (CDM). The paper details the architectural framework of the Internet of Medical Things (IoMT), which relies on wearable and implantable devices to continuously acquire physiological and biochemical data from patients. This raw data is transformed into actionable intelligence using AI contributions, notably Machine Learning (ML) and Deep Learning (DL) algorithms, which are essential for identifying, classifying, and forecasting diseases like diabetes and heart conditions. The system operates across four layers: data acquisition via sensors, secure data transmission using protocols like Bluetooth and Wi-Fi, data processing (often leveraging Edge Computing to minimize latency), and a user interface for timely clinical decision-making. Standard protocols such as HL7 and FHIR are highlighted as crucial for ensuring secure and scalable interoperability. The review also critically assesses the significant challenges confronting widespread adoption, including data privacy and security, the complexity of device interoperability, scalability issues related to massive data volumes, and ethical concerns around AI bias and transparency. Future research is encouraged to explore emerging technologies like Federated Learning and blockchain for enhanced data privacy, the development of Explainable AI (XAI), and advancements in energy-efficient sensor technologies.

Key words: Internet of Things (IoT), Artificial Intelligence (AI), Biomedical Sensors, Health Risk Prediction, Chronic Disease Management (CDM), Internet of Medical Things (IoMT), Machine Learning (ML), Deep Learning (DL), Data Privacy and Security, Interoperability, Explainable AI (XAI), Federated Learning, Edge Computing, Healthcare