Smart-Watches Assisted Sugar Level Monitoring with Different Activities and Nutrition based on Machine Learning Approaches

Abstract

These days, sugar glucose monitoring is very important for both diabetic and non-diabetic patients while they are eating and doing different activities in practice. There are different ways to monitor body glucose levels such as blood-based glucose monitoring and smart watches-based glucose monitoring. However, continuous glucose monitoring (CGM) is an emerging non-invasive method for different subjects (e.g., patients and customers). However, smartwatches have limitations. In this paper, we present a new smartwatch framework that monitors the body's glucose level with new features such as nutrition, and activities. We present the modified dataset with an additional feature such as sugar glucose level with different activities (e.g., running, sitting, sleeping, and walking) while eating different nutrition in different time intervals. We present empirical machine learning such as an activity glucose monitoring algorithm (ASA) which executes all datasets with more optimal results. Simulation results show that our proposed framework is more optimal and shows glucose monitoring with different activities with more features as compared to existing smartwatches and obtained an accuracy of 78% as compared to existing machine learning methods.