To improve the accuracy of predicting non-invasive blood glucose concentration in the near-infrared spectrum, we utilized the Particle Swarm Optimization (PSO) algorithm to optimize hyperparameters for the Multi-Kerne...To improve the accuracy of predicting non-invasive blood glucose concentration in the near-infrared spectrum, we utilized the Particle Swarm Optimization (PSO) algorithm to optimize hyperparameters for the Multi-Kernel Learning Support Vector Machine (MKL-SVR). With these optimized hyperparameters, we established a non-invasive blood glucose regression model, referred to as the PSO-MKL-SVR model. Subsequently, we conducted a comparative analysis between the PSO-MKL-SVR model and the PSO-SVR model. In a dataset comprising ten volunteers, the PSO-MKL-SVR model exhibited significant precision improvements, including a 16.03% reduction in Mean Square Error and a 0.29% increase in the Squared Correlation Coefficient. Moreover, there was a 0.14% higher probability of the Clark’s Error Grid Analysis falling within Zone A. Additionally, the PSO-MKL-SVR model demonstrated a faster operational speed compared to the PSO-SVR model.展开更多
Diabetes mellitus is a huge and significantly grow ing problem. Continuous and real-time monitoring of blood glucose plays a key role for the people with diabetes,which can help them to control glucose concentration m...Diabetes mellitus is a huge and significantly grow ing problem. Continuous and real-time monitoring of blood glucose plays a key role for the people with diabetes,which can help them to control glucose concentration more effectively. However,current blood glucose monitoring methods require blood by needle-pricking,which limit the detection frequency. It is necessary to develop non-invasive blood glucose monitoring methods to achieve the ideal therapeutic and management of diabetes. In this paper,the developments and challenges of non-invasive blood glucose monitoring technologies in recent years are reviewed. And the bottleneck and the developing trends are also analyzed.展开更多
A new non-invasive blood glucose measuring apparatus (NBGMA) made up of MSP430F149 SCM (single chip micyoco) was developed,which can measure blood glucose level (BGL) frequently,conveniently and painlessly. The hardwa...A new non-invasive blood glucose measuring apparatus (NBGMA) made up of MSP430F149 SCM (single chip micyoco) was developed,which can measure blood glucose level (BGL) frequently,conveniently and painlessly. The hardware and software of this apparatus were designed,and detecting algorithms based on conservation of energy method (COEM) were presented. According to the law of conservation of energy that the energy derived by human body equals energy consumed by metabolism,and the relationship between convection,evaporation,radiation and the BGL was established. The sensor module was designed. 20 healthy volunteers were involved in the clinical experiment. The BGL measured by an automatic biochemical analyzer (ABA) was set as the reference. Regression analysis was performed to compare the conservation of energy method with the biochemical method,using the 20 data points with blood glucose concentrations ranging from 680 to 1 100 mg/L. Reproducibility was measured for healthy fasting volunteers. The results show that the means of BGL detected by NBGMA and ANA are very close to each other,and the difference of standard deviation (SD) is 24.7 mg/L. The correlative coefficient is 0.807. The coefficient of variation (CV) is 4% at 921.6 mg/L. The resultant regression is evaluated by the Clarke error grid analysis (EGA) and all data points are included in the clinically acceptable regions (region A:100%,region B:0%). Accordingly,it is feasible to measure BGL with COEM.展开更多
Diabetes mellitus is a group of metabolic disorders of carbohydrate metabolism in which glucose is underutilized, producing hyperglycaemia. The management requires proper maintenance of glucose and electrolytes within...Diabetes mellitus is a group of metabolic disorders of carbohydrate metabolism in which glucose is underutilized, producing hyperglycaemia. The management requires proper maintenance of glucose and electrolytes within its optimum concentrations. The study was designed to evaluate the possibility of using saliva as an alternative non-invasive sample for the determination of electrolytes. A total of 100 samples were utilized consisting of equal number of control (non-diabetics) and diabetic groups. Fasting blood and saliva were collected employing standard methods. The biochemical parameters were analysed using WHO approved methods and procedures. Independent samples t-test and Pearson correlation were the statistical tools used for the data analysis obtained from SPSS package (version 20). The study revealed a significant increase (p < 0.05) in concentrations of blood and salivary glucose, potassium and calcium when controls were compared to diabetics. Moreover, there was a high level of semblances and patterns between plasma and salivary electrolytes, except for potassium. Therefore, electrolytes and glucose results gotten from the use of saliva could be used to equate to that of blood. Hence, instances of non-accessibility of blood, saliva could be of help.展开更多
文摘To improve the accuracy of predicting non-invasive blood glucose concentration in the near-infrared spectrum, we utilized the Particle Swarm Optimization (PSO) algorithm to optimize hyperparameters for the Multi-Kernel Learning Support Vector Machine (MKL-SVR). With these optimized hyperparameters, we established a non-invasive blood glucose regression model, referred to as the PSO-MKL-SVR model. Subsequently, we conducted a comparative analysis between the PSO-MKL-SVR model and the PSO-SVR model. In a dataset comprising ten volunteers, the PSO-MKL-SVR model exhibited significant precision improvements, including a 16.03% reduction in Mean Square Error and a 0.29% increase in the Squared Correlation Coefficient. Moreover, there was a 0.14% higher probability of the Clark’s Error Grid Analysis falling within Zone A. Additionally, the PSO-MKL-SVR model demonstrated a faster operational speed compared to the PSO-SVR model.
基金supported by the National Natural Science Foundation of China (No.81471698 and 81401454)the National High Technology Research and Development Program of China (‘863’ Program,No.2012AA022602)
文摘Diabetes mellitus is a huge and significantly grow ing problem. Continuous and real-time monitoring of blood glucose plays a key role for the people with diabetes,which can help them to control glucose concentration more effectively. However,current blood glucose monitoring methods require blood by needle-pricking,which limit the detection frequency. It is necessary to develop non-invasive blood glucose monitoring methods to achieve the ideal therapeutic and management of diabetes. In this paper,the developments and challenges of non-invasive blood glucose monitoring technologies in recent years are reviewed. And the bottleneck and the developing trends are also analyzed.
基金Project(07JJ6133) supported by the Natural Science Foundation of Hunan Province, China
文摘A new non-invasive blood glucose measuring apparatus (NBGMA) made up of MSP430F149 SCM (single chip micyoco) was developed,which can measure blood glucose level (BGL) frequently,conveniently and painlessly. The hardware and software of this apparatus were designed,and detecting algorithms based on conservation of energy method (COEM) were presented. According to the law of conservation of energy that the energy derived by human body equals energy consumed by metabolism,and the relationship between convection,evaporation,radiation and the BGL was established. The sensor module was designed. 20 healthy volunteers were involved in the clinical experiment. The BGL measured by an automatic biochemical analyzer (ABA) was set as the reference. Regression analysis was performed to compare the conservation of energy method with the biochemical method,using the 20 data points with blood glucose concentrations ranging from 680 to 1 100 mg/L. Reproducibility was measured for healthy fasting volunteers. The results show that the means of BGL detected by NBGMA and ANA are very close to each other,and the difference of standard deviation (SD) is 24.7 mg/L. The correlative coefficient is 0.807. The coefficient of variation (CV) is 4% at 921.6 mg/L. The resultant regression is evaluated by the Clarke error grid analysis (EGA) and all data points are included in the clinically acceptable regions (region A:100%,region B:0%). Accordingly,it is feasible to measure BGL with COEM.
文摘Diabetes mellitus is a group of metabolic disorders of carbohydrate metabolism in which glucose is underutilized, producing hyperglycaemia. The management requires proper maintenance of glucose and electrolytes within its optimum concentrations. The study was designed to evaluate the possibility of using saliva as an alternative non-invasive sample for the determination of electrolytes. A total of 100 samples were utilized consisting of equal number of control (non-diabetics) and diabetic groups. Fasting blood and saliva were collected employing standard methods. The biochemical parameters were analysed using WHO approved methods and procedures. Independent samples t-test and Pearson correlation were the statistical tools used for the data analysis obtained from SPSS package (version 20). The study revealed a significant increase (p < 0.05) in concentrations of blood and salivary glucose, potassium and calcium when controls were compared to diabetics. Moreover, there was a high level of semblances and patterns between plasma and salivary electrolytes, except for potassium. Therefore, electrolytes and glucose results gotten from the use of saliva could be used to equate to that of blood. Hence, instances of non-accessibility of blood, saliva could be of help.