⁠Effectiveness of Non-Invasive Sensor-Based Tools for Blood Glucose Detection

Abstract

Background: Monitoring blood glucose levels is one of the main pillars of diabetes management to prevent complications and reduce the risk of morbidity and mortality. Today's blood glucose monitoring is a non-invasive method that offers speed, accuracy, and painless convenience. Referring to this need, this study aims to demonstrate the effectiveness of non-invasive sensor-based detection devices in checking blood glucose levels in order to provide a more comfortable and efficient alternative for diabetes patients. Methods: This study developed a non-invasive glucometer using the latest and smaller version of Arduino Uno and tested it on 20 samples, consisting of 10 diabetes mellitus patients and 10 with normal blood glucose. The test was carried out by comparing the measurement results from the non-invasive device and the standard GCU Easy Touch 3-in-1 device to determine the accuracy of the device. The tool-testing method uses sensitivity, specificity, and accuracy. Results: This non-invasive measuring tool is more effective when used to measure patients with diabetes mellitus. This device shows an error rate of 9.21%, a sensitivity of 80%, and a specificity of 50%. Meanwhile, the overall measurement accuracy, calculated at 83.3%, reinforces the tool's effectiveness in providing reliable results. Conclusion: This device has the potential to be a convenient and painless method of blood glucose monitoring for diabetic patients. However, further development is needed to improve the development of machine learning-based algorithms to process sensor data so that tools can identify unique patterns from each individual and provide more accurate results.