Study Of Glucolipid Metabolism Assessment Method Based Respiratory Quotient Assay

Diabetes is one of the most common lifelong chronic diseases for human,and diabetic patients are often accompanied by disorders of glucolipid metabolism.Respiratory Quotient(RQ)can reflect the ratio of respiratory substrate utilization in the body metabolic process.This paper deeply investigates the method of glucolipid metabolism assessment based on respiratory quotient detection,which is important for non-invasive screening and monitoring of diabetes.In this paper,a portable non-protein Respiratory Quotient(np RQ)detection system based on STM32 is developed.The hardware circuit part of the detection system includes a microcontroller minimum system circuit,a display module circuit,a power supply module circuit,a signal acquisition module circuit and a microcontroller communication circuit.The software part uses C programming language to write the reading program of each sensor signal through MDK5,and uses the specific algorithm of np RQ to calculate the np RQ and displays its results on the OLED screen.Finally,a bivariate Pearson correlation analysis is performed between the results of the developed portable non-protein respiratory quotient testing system and the results of the U.S.Sendis metabolic cart,and the correlation coefficient reached 0.978.So the portable system is highly accurate and stable in operation.Besides,this paper designs specific Oral Glucose Tolerance Test(OGTT)and bun meal test for healthy volunteers with normal glucolipid metabolism and type II diabetic patients with disordered glucolipid metabolism accordingly,while the portable nonprotein respiratory quotient testing system is used to detect the subjects’ np RQ.Through the correlation analysis of physiological parameters such as np RQ and blood glucose value,the correlation coefficient between np RQ and blood glucose value was obtained over 0.8.Further screening and analysis of the experimental data show that there are significant differences between the data of volunteers with normal glucolipid metabolism and diabetic patients with disordered glucolipid metabolism at 2 hours after lunch in the diet intervention experiment.Therefore,the experimental data at this time is used as the data set,and the Body Mass Index(BMI)index was added as the input with np RQ to establish two blood glucose prediction models which is based on multiple linear regression and BP neural network.The performance evaluation indexes of the models shows that both models have good prediction performance,and the correlation coefficients reach 0.878 and 0.912,respectively.With the real blood glucose value and predicted blood glucose value as the horizontal and vertical coordinates,a Clark error network analysis graph is established,in which 90% of the coordinate points fell into the A area of the Clark error raster.By assessing the predicted blood glucose values of2 hours after meal in the dietary intervention experiment,this paper finally provides a new method for assessing the body’s glucolipid metabolism.In conclusion,the study of glucolipid metabolism assessment based on respiratory quotient testing has important clinical significance for the future assessment of the risk of diseases such as diabetes,and can also better assist in non-invasive blood glucose measurement.