Non-Invasive Blood Glucose Estimation System Based on a Neural Network with Dual-Wavelength Photoplethysmography and Bioelectrical Impedance Measuring

Abstract

This study proposed a noninvasive blood glucose estimation system based on dual-wavelength photoplethysmography (PPG) and bioelectrical impedance measuring technology that can avoid the discomfort created by conventional invasive blood glucose measurement methods while accurately estimating blood glucose. The measured PPG signals are converted into mean, variance, skewness, kurtosis, standard deviation, and information entropy. The data obtained by bioelectrical impedance measuring consist of the real part, imaginary part, phase, and amplitude size of 11 types of frequencies, which are converted into features through principal component analyses. After combining the input of seven physiological features, the blood glucose value is finally obtained as the input of the back-propagation neural network (BPNN). To confirm the robustness of the system operation, this study collected data from 40 volunteers and established a database. From the experimental results, the system has a mean squared error of 40.736, a root mean squared error of 6.3824, a mean absolute error of 5.0896, a mean absolute relative difference of 4.4321%, and a coefficient of determination (R Squared, R2) of 0.997, all of which fall within the clinically accurate region A in the Clarke error grid analyses.

Non-Invasive Glucose Sensing Technologies and Products: A Comprehensive Review for Researchers and Clinicians

“…However, the researchers stated that further research is needed to understand the ideal frequency range for measuring glucose levels. Similarly, Yen, Chen [41] demonstrated improved accuracy with the proposed combination of dual-wavelength PPG and BIS in the frequency range of 50 and 100 kHz placed between the index and middle fingers as well as under the wrists while using a Back Propagation Neural Network algorithm. However, it is important to consider the wrist circumference and wrist temperature, as well as body weight, as physiological parameters that can impact the accuracy of non-invasive glucose measurements [50].…”

“…Yen, Chen [41] found that combining dual-wavelength PPG with another method, such as Bioelectrical Impedance (BI), delivered quantitative results. Although BI does not directly measure glucose levels, it can provide information about body composition and fluid distribution (i.e., body fat, mass, and water).…”

“…However, the re searchers stated that further research is needed to understand the ideal frequency range for measuring glucose levels. Similarly, Yen, Chen [41] demonstrated improved accuracy Using Bio-Electromagnetic Resonance (BEMR), GlucoBand by Calisto Medical (Plano, TX, USA) has previously undergone trials for a continuous wristwatch connected to an armband; however, it has disappeared from the market [12]. Pedro, Marcôndes [122] used frequency ranges between 50 and 70 kHz alongside Effective Medium Theory (EMT) machine learning and found the estimated changes in glucose levels.…”

Accuracy of a Non-Invasive Home Glucose Monitor for Measurement of Blood Glucose

“…Blood glucose has different absorption rates of different lights and these data are used to calculate the blood glucose. [ 17 ] A previous study showed that GlucoTrack can detect blood glucose from the ear lobe with clinical accuracy. However, this device is not available to Indian consumers.…”