Machine edge-aware IoT framework for real-time health monitoring: Sensor fusion and AI-driven emergency response in decentralized networks

Abstract

Health monitoring systems require wider deployment to support medical institutions in their care of both chronic patients and elderly people while providing urgent emergency services. Systems based on traditional cloud infrastructure through central locations create multiple problems that include delays in performance along with failures in connectivity and restrictions in system usage, and privacy risks. The IoT framework resolves issues through real-time sensor fusion, where edge-based decision systems use lightweight AI anomaly detectors for urgent emergency choices. Several wearable biosensors analyze heart rate and blood oxygen saturation rates and body temperature information as they process fall metrics live. The edge nodes perform speedy AI-based analytics directly for essential health situations through their efficient processing capabilities to establish brief cloud system connections. In distributed networks, system-wide alerts are activated through the emergency alert protocol after distributed networks execute autonomous decision-making processes. The proposed design utilizes an adaptive mesh networking approach to ensure dependable transmission across diverse settings and support ongoing remote monitoring. A hybrid sensor fusion algorithm analyses different physiological parameters, such as ECG, SpO₂, and body temperature, to detect potentially dangerous signals and set off local emergency alarms. The figures show that the system achieves high detection accuracy (up to 95.4 %) within just 0.045 s and uses less power. The efficacy of the fall and cardiac event detection capabilities was shown in a simulation of real-life assisted living settings. The findings demonstrate that the system provides accurate, secure health data sharing in a scalable and rapid manner for dispersed IoT environments.