Research And Design Of Noninvasive Blood Glucose Detection System

With the rapid development of China's social economy,the increase of work pressure has led to more and more overtime workers.The poor schedule and diet structure have caused their bodies to be in a sub-health state.More and more people suffer from chronic diseases.The development of diabetes is particularly rapid,and it has become one of the three major diseases in the world that endanger human health.One of the characteristics of diabetes is the increase in blood glucose concentration.Therefore,real-time detection of human blood glucose concentration is very important.At present,the commonly used blood glucose testing instruments are invasive or minimally invasive,which will cause the patient to have physical and psychological burden and increase the risk of unnecessary infection.Therefore,in order to meet the patient's painless real-time monitoring of blood glucose requirements,a non-invasive blood glucose testing instrument is needed by the majority of patients.Based on the theory of near-infrared spectroscopy and energy metabolism conservation,this paper establishes a non-invasive blood glucose detection system model and proposes a novel non-invasive blood glucose algorithm based on RNN neural network.This article is mainly composed of the following parts:1.Research on the theoretical basis of non-invasive blood glucose testing.This paper mainly introduces the principle of photoelectric method for measuring blood sugar--Lambert Beer's law,and deduces the blood glucose equation.Because of the low precision caused by human body model and systematic error,it is proposed to use the energy metabolism conservation method to correct the blood sugar level,and then measure The process proposes requirements,including the selection of correction parameters and the selection of detection locations,methods,wavelengths,and so on.2.Design the hardware and software of the non-invasive blood glucose detection system.The hardware includes embedded microprocessor module,sensor module,information processing module and network communication module.The software module mainly includes filtering,feature extraction,blood glucose correction and other algorithms and C#-based visual upper computer design.3.Research on the algorithm of non-invasive blood glucose detection system.Firstly,the MCU uploads the information to the host computer through the network communication module to perform filtering and separation work,then introduces several characteristic parameters of the human body,then introduces four different blood glucose correction algorithms,and acquires multiple human body features through the system device.The value,the composition of the characteristic matrix,combined with the blood glucose concentration reference value measured by the invasive blood glucose meter for data analysis.4.Experimental verification.Select the appropriate experimental environment and experimental equipment,carry out the data model simulation experiment in MATLAB,select the optimal blood glucose correction algorithm model and write it in the upper computer.5.Put forward the shortcomings and shortcomings of this paper,and put forward the next research direction.