Noninvasive Continuous Glucose Monitoring Using Multimodal Near-Infrared, Temperature, and Pressure Signals on the...

Area of Science:

• Biomedical Engineering
• Optical Sensing Technologies
• Medical Diagnostics

Background:

• Continuous glucose monitoring (CGM) is crucial for diabetes management.
• Existing invasive methods pose challenges like discomfort and infection risk.
• Noninvasive glucose monitoring (NIGM) remains a significant unmet need.

Purpose of the Study:

• To develop and validate a novel noninvasive continuous glucose monitoring (NI-CGM) system for earlobe application.
• To enhance the reliability of NIGM using multimodal sensing and advanced spectroscopy.
• To assess the system's accuracy and feasibility in a pilot clinical trial.

Main Methods:

• Utilized earlobe anatomy for optimal optical transmission in the NI-CGM system.
• Integrated near-infrared (NIR) diffuse transmission, temperature, and pressure sensors.
• Developed a Multi-Wavelength Slope Efficiency Near-Infrared Spectroscopy (MW-SE-NIRS) method for improved noise robustness.
• Employed a 1D convolutional neural network (Conv1D) for data analysis.

Main Results:

• Achieved 97.0% accuracy in the A-Zone of the Clarke Error Grid (CEG) and 5.2% MARD in stable physiological conditions for three subjects.
• Demonstrated 90.9% CEG A-Zone accuracy and 8.4% MARD across all five participants.
• Identified individual factors like earlobe thickness and physical activity influencing performance.

Conclusions:

• The MW-SE-NIRS based earlobe NI-CGM system shows significant potential for accurate, noninvasive glucose monitoring.
• Further research is needed for personalized modeling and larger clinical validation.
• This technology could offer a comfortable and effective alternative to invasive glucose sensors.