IoT-based Smart Wearable Analysis for Personalised Healthcare using 5G-Assisted Machine Learning

Abstract

Over the last few years, the Internet of Things (IoT) has drastically transformed the healthcare industry by enabling real-time monitoring services using smart wearable devices like smartwatches, rings, etc. IoT-based smart wearable devices are changing traditional processes of healthcare to personalized healthcare systems. It enhances persons daily activities, boosts people's well-being, and transforms our quality of living (QoL). Further, smart wearable devices help in monitoring and improving personalized healthcare by tracking day-to-day activities, where IoT-based sensors collect data from smart wearables and stored it in a cloud-based storage system. From the cloud, data is further gathered for analysis. Several research work has been done so far in this regard but it has not been exploited fully. Hence, this study proposed a machine learning (ML)-based digital healthcare system for the precise prediction of daily activities using real-time data. Next, experimental results are evaluated using ML models like Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF) applied for the activity prediction. The proposed approach proved its effectiveness by comparing it with the traditional system with respect to various performance evolution matrices of accuracy, root mean square error (RMSE), mean squared error (MSE), mean absolute error (MAE), and R2 score.