Search

Blood glucose monitoring is vital for controlling the complications of diabetes. We propose a noninvasive blood glucose monitor featuring multi-sensor fusion, which functions well for rapid yet accurate blood glucose monitoring and screening of diabetes. The characteristic parameters of diabetics were extracted by applying one-factor analysis of variance, and a classification model was established via principal components analysis and discriminant analysis to identify whether the subjects suffer from diabetes. In light of the challenges for noninvasive blood glucose monitoring modelling brought by different pathogeneses, disease processes and complicated complications of different diabetics, the subjects were classified into different categories via unsupervised K-means clustering algorithm and modelled for each category separately for expanded coverage and improved accuracy. The accuracy was clinically evaluated in 254 volunteers with type II diabetes. The results in the Parkes error grid were as follows: 58.33% in Zone A, 39.43% in Zone B and 2.24% in Zone C, with a correlation coefficient of 0.69 and a root mean squared error of 2.67 mmol/L. In 2019, the monitor was certified by the National Medical Products Administration of China as the first Class III Medical Device.

Daily continuous glucose monitoring is very helpful in the control of glucose levels for people with diabetes and impaired glucose tolerance. In this study, a multisensor-based, noninvasive continuous glucometer was developed, which can continuously estimate glucose levels via monitoring of physiological parameter changes such as impedance spectroscopy at low and high frequency, optical properties, temperature and humidity. Thirty-three experiments were conducted for six healthy volunteers and three volunteers with diabetes. Results showed that the average correlation coefficient between the estimated glucose profiles and reference glucose profiles reached 0.8314, with a normalized root mean squared error (NRMSE) of 14.6064. The peak time of postprandial glucose was extracted from the glucose profile, and its estimated value had a correlation coefficient of 0.9449 with the reference value, wherein the root mean square error (RMSE) was 6.8958 min. Using Clarke error grid (CEG) analysis, 100% of the estimated glucose values fell in the clinically acceptable zones A and B, and 92.86% fell in zone A. The application of a multisensor-based, noninvasive continuous glucometer and time series analysis can endure the time delay between human physiological parameters and glucose level changes, so as to potentially accomplish noninvasive daily continuous glucose monitoring.