IoT-Based Blood Glucose Detection System

This study presents a non-invasive blood glucose concentration (BGC) system that is based on the Internet of Things (IoT) and employs near-infrared sensors (NIR) to detect blood glucose levels through the scanning of fingers. The purpose is to develop a BGC system and assess the accuracy of the proposed system. Constant monitoring and data-driven insights for optimum treatment and enhanced quality of life are made possible by IoT-based systems, and non-invasive glucose monitoring is a critical component of diabetes care. The system that was built comprises of a Light Emitting Diode (LED) that emits signals that are sent via the fingertip, and a phototransistor that is positioned next to the LED to detect any reflected signals. The results of the scanning are then stored in ThingSpeak. The technique used to determine the blood glucose level involves analyzing the variation in received signal intensity that results from reflection. Using regression analysis, a mathematical relationship between glucose concentration and voltage was established and installed in the Arduino. Two sorts of testing were carried out to evaluate the system: in vitro tests on glucose solution and in vivo experiments on the human body. By comparing 10 readings acquired from both types of experiments, the results of the studies indicate that the device's glucose detection accuracy ranges from 1.13 to 16.41%. IoT approach is feasible due to its continuous monitoring capabilities, data analysis, and remote access features, making it a promising tool for effective diabetes control.