Quantitative measurement of blood glucose influenced by multiple factors via photoacoustic technique combined with optimized wavelet neural networks

Abstract

In this work, the photoacoustic (PA) quantitative measurement of blood glucose concentration (BGC) influenced by multiple factors was firstly investigated. A set of PA detection system of blood glucose considering the comprehensive influence of five factors was established. The PA signals and peak‐to‐peak values (PPVs) of 625 rabbit whole blood were obtained under 625 influence combinations. Due to the accurate measurement of BGC limited by the overlap PA signals, wavelet neural network (WNN) was utilized to train the PPVs of blood glucose for 500 rabbit blood. The mean square error (MSE) of BGC for 125 testing blood was approximately 6.5782 mmol/L. To decrease the MSE, the parameters of WNN were optimized by particle swarm optimization (PSO), that is, PSO‐WNN algorithm was employed. Under the optimal parameters, MSE of BGC was decreased to approximately 0.48005 mmol/L. To further improve the prediction accuracy of BGC, an improved nonlinear dynamic inertia weight (NDIW) strategy of PSO was proposed, and compared with other two kinds of dynamic inertia weight strategies. Under the optimal parameters, the MSE of BGC was decreased to approximately 0.2635 mmol/L. The comparison of nine algorithms demonstrate that the PA technique combined with PSO‐WNN and the improved NDIW strategy is significant in the quantitative measurement of blood glucose influenced by multiple factors.

Quantitative blood glucose detection influenced by various factors based on the fusion of photoacoustic temporal spectroscopy with deep convolutional neural networks

“…After 500 times training cycles for the training set samples, 125 randomly selected samples from the testing set were input into the trained 1DCNN-SAM-LSTM model. The results demonstrate that 1DCNN-SAM-LSTM achieved the superior performance with MSE C of 0.034317 mmol/L on the training set and MSE P of 0.14320 mmol/L on the testing set when NumberHeads is 32 and NumChannels of SAM module is 128, which is better than that of the previous results [ 17 , 20 ]. The clarke error grid graph of BGC for 125 testing set samples with the synthetical influences of various factors is depicted in Fig.…”

“…At the same time, the external factors affecting the photoacoustic signal, such as laser energy, liquid temperature and detection distance were explored, and a calibration method was also proposed. Ren [ 16 , 17 ] investigated the single factor and multiple factors influences of blood glucose photoacoustic detection for the glucose solution and animal whole blood. The positive relationship between the photoacoustic intensity and the laser energy, temperature and glucose concentration were obtained, as well as the negative relationship between the photoacoustic intensity and the flow velocity and detection distance.…”

A Study on Near-Infrared Non-Invasive Blood Glucose Concentration Regression Prediction Based on PSO-MKL-SVR

“…Zhang et al [9] proposed a method that combines Particle Swarm Optimization with the Generalized Regression Neural Network (GRNN), showcasing impressive predictive performance with a root mean square error of 0.26 in non-invasive blood glucose detection. Ren et al [10] introduced an enhanced PSO combined with the Wavelet Neural Network, improving the accuracy of blood glucose concentration prediction. Additionally, Li et al [11] This paper employs the PSO-MKL-Support Vector Machine Regression (SVR) method to construct a non-invasive near-infrared blood glucose prediction model.…”