IOT-INTEGRATED PREDICTIVE HEALTHCARE SYSTEMS FOR ELDERLY PATIENT MONITORING USING MACHINE LEARNING AND CLOUD COMPUTING

Abstract

The rapid growth of aging populations has increased the demand for intelligent healthcare systems capable of providing continuous monitoring, early disease detection, and proactive medical intervention. Traditional healthcare approaches often rely on periodic clinical assessments, which may fail to identify sudden physiological abnormalities among elderly patients. This research proposes an IoT-Integrated Predictive Healthcare System for Elderly Patient Monitoring using Machine Learning and Cloud Computing to enhance remote healthcare management and health-risk prediction. The proposed framework combines wearable IoT sensors, cloud-based healthcare infrastructure, and machine learning analytics to continuously monitor physiological and behavioural health indicators. A healthcare dataset containing 5,000 elderly patient records was utilized, incorporating parameters such as heart rate, blood pressure, oxygen saturation, respiratory rate, glucose level, body temperature, body mass index, sleep duration, activity level, ECG abnormality score, and fall-detection status. Data preprocessing, feature extraction, and feature-importance analysis were performed before predictive model development. Random Forest, Gradient Boosting, and Voting Ensemble classifiers were implemented for health-risk classification. Experimental results demonstrated that the ensemble-learning framework achieved superior predictive performance compared with individual machine learning models. Feature-importance analysis identified cardiovascular indicators, glucose measurements, oxygen saturation levels, and ECG abnormality scores as major contributors to healthcare-risk prediction. The proposed framework supports real-time healthcare monitoring, automated alert generation, cloud-based healthcare management, and intelligent decision support for caregivers and healthcare professionals. The integration of IoT, machine learning, and cloud computing enables scalable, efficient, and proactive elderly healthcare management. The findings confirm that predictive healthcare systems can significantly improve healthcare accessibility, patient safety, remote monitoring capability, and early health-risk identification in modern smart healthcare environments.