Noninvasive Continuous Glucose Monitoring Using a Multisensor-Based Glucometer and Time Series Analysis

Area of Science:

• Biomedical Engineering
• Medical Devices
• Physiological Monitoring

Background:

• Daily continuous glucose monitoring is crucial for managing diabetes and impaired glucose tolerance.
• Current monitoring methods can be invasive or lack continuous data.
• Accurate, noninvasive glucose estimation is a significant unmet clinical need.

Purpose of the Study:

• To develop and evaluate a multisensor-based, noninvasive continuous glucometer.
• To assess the device's ability to estimate glucose levels by monitoring physiological parameters.
• To determine the clinical accuracy and reliability of the noninvasive glucose monitoring system.

Main Methods:

• Development of a noninvasive glucometer utilizing impedance spectroscopy (low and high frequency), optical properties, temperature, and humidity sensors.
• Conducting 33 experiments on six healthy and three diabetic volunteers.
• Employing time series analysis to account for physiological parameter and glucose level time delays.

Main Results:

• Achieved an average correlation coefficient of 0.8314 between estimated and reference glucose profiles.
• Reported a normalized root mean squared error (NRMSE) of 14.6064 for glucose estimation.
• 100% of estimated glucose values fell within clinically acceptable zones (A and B) on the Clarke Error Grid analysis, with 92.86% in zone A.

Conclusions:

• The developed multisensor-based, noninvasive continuous glucometer demonstrates significant potential for accurate glucose level estimation.
• The system effectively addresses the time delay between physiological parameter changes and actual glucose fluctuations.
• This technology offers a promising noninvasive alternative for daily continuous glucose monitoring in diabetes care.