Real-Time Smart Healthcare System Based on Edge-Internet of Things and Deep Neural Networks for Heart Disease Prediction

Abstract

With technological advancements, smart health monitoring systems have become increasingly vital and popular. The rise of smart homes, appliances, and medical systems, along with the pivotal role of the Internet of Things (IoT), is significantly enhancing healthcare services by improving data processing and predictive capabilities. IoT not only aids in predicting heart disease but also supports emergency responses. However, traditional data transfer methods are inefficient in terms of time and energy, resulting in high latency and consumption. Edge computing, alongside deep learning methods, provides effective solutions with superior performance. This paper introduces a Real-Time Smart Healthcare System utilizing Edge-Internet of Things and Deep Learning. The primary objective of this system is to monitor patient health changes, predict heart disease, and automate medication administration in real time. The study presents a DNN-based prediction model that leverages edge computing and IoT. This model processes health data from IoT devices, while edge devices deliver timely health predictions to doctors and patients via edge and cloud servers. The proposed system is evaluated on performance parameters, demonstrating superior results compared to other methods. By integrating edge computing, IoT, and deep learning, this system enables efficient real-time health monitoring and prediction, benefiting both healthcare professionals and patients. It demonstrates exceptional performance with an accuracy of 96.15%, precision of 92.86%, recall (sensitivity) of 97.50%, and an F1-score of 94.87%.