A SMART HEALTHCARE SURVEILLANCE ARCHITECTURE INTEGRATING IOT-BASED SENSING AND MACHINE LEARNING LEVERAGING FOR INTELLIGENT PATIENT SURVEILLANCE

Abstract

The integration of Artificial Intelligence (AI) with the Internet of Things (IoT) is creating new opportunities to improve healthcare services through continuous patient monitoring, timely medical assessment, and data-driven decision-making. This study proposes an intelligent remote health monitoring framework that combines IoT-enabled sensors with machine learning techniques to support efficient, reliable, and personalized healthcare. The proposed approach focuses on improving the accuracy of health assessments, enabling real-time analysis of physiological data, protecting sensitive patient information, and facilitating effective communication between system components. By addressing these requirements, the framework aims to support early identification of potential health concerns and more proactive patient care. The proposed system employs machine learning algorithms, including Random Forest (RF), Support Vector Machine (SVM), and K-Nearest Neighbors (KNN), to examine physiological parameters such as heart rate, blood pressure, and electrocardiogram (ECG) signals. Particular attention is given to challenges involving the reliability of collected health data, patient privacy, and the computational demands of continuous monitoring. Experimental evaluations indicate that Random Forest achieves better overall classification performance than SVM and KNN, demonstrating stronger results in accuracy, precision, and recall within the evaluated setting. These findings suggest that combining IoT-based data collection with machine learning can provide valuable support for remote health assessment and timely clinical decision-making. The proposed framework also offers a basis for developing accessible, secure, and scalable healthcare technologies that can adapt to individual patient needs. Future development may further strengthen its capabilities through enhanced privacy mechanisms, efficient data processing, and validation using larger and more diverse clinical datasets.