First Experiences With a Wearable Multisensor Device in a Noninvasive Continuous Glucose Monitoring Study at Home,...

Area of Science:

• Biomedical Engineering
• Endocrinology
• Medical Devices

Background:

• Previous studies demonstrated the utility of noninvasive continuous glucose monitoring (CGM) with wearable multisensor devices for diabetes therapy in controlled settings.
• Wearable CGM devices offer valuable insights into glucose trends, aiding in diabetes management.

Purpose of the Study:

• To evaluate the effectiveness of a noninvasive wearable multisensor device for continuous glucose monitoring in uncontrolled, real-world diabetes conditions.
• To assess the performance of an established algorithmic routine in estimating glucose levels noninvasively outside of controlled environments.

Main Methods:

• A long-term, at-home study involved 20 individuals with type 1 diabetes, collecting 1072 study days of multisensor and reference glucose data.
• A fully online-compatible algorithmic routine was employed to link multisensor data to glucose levels for noninvasive estimation.
• The algorithm incorporated daily calibration to calculate glucose values from sensor data.

Main Results:

• The developed algorithm achieved a Mean Absolute Relative Difference (MARD) of 35.4 mg/dL on test data.
• While less accurate than minimally invasive devices, 86.9% of glucose rate points fell within the acceptable AR+BR zone.
• The multisensor device and algorithm successfully tracked glucose changes in uncontrolled conditions, albeit with reduced accuracy.

Conclusions:

• The wearable multisensor device and algorithmic routine can track glucose changes in uncontrolled diabetes settings, though accuracy is lower than in controlled environments.
• Analysis of learning curves indicated that further data collection alone is unlikely to significantly improve current results.
• Future research should prioritize developing more sophisticated algorithms to better address environmental and physiological confounding factors affecting glucose monitoring accuracy.