DefinePK

Abstract

Remote patient monitoring (RPM) has gained momentum with the proliferation of Internet of Things (IoT) devices and advancements in machine learning (ML). This research proposes an IoT-enabled RPM system integrated with ML models to enable early disease prediction and health trend analysis. The system collects real-time physiological data from wearable devices and environmental sensors and employs supervised learning algorithms for anomaly detection and risk classification. Our experiments conducted using a synthesized dataset simulating real-world vitals (e.g., heart rate, oxygen saturation, temperature), show that models like Random Forest and LSTM can predict critical health conditions with over 93% accuracy. This paper highlights the architecture, data pipeline, and predictive capabilities of the system, underscoring its potential in reducing hospital readmissions and enabling proactive healthcare