Design And Implementation Of Multi-Physiological Parameter Monitoring System

The health problems of the elderly brought about by the ever-increasing population aging have become a major problem facing the society.At the same time,cardiovascular diseases,which are more harmful to the health of the elderly,also show a trend of younger patients.So,it is imperative to improve the physical quality of the people and solve the problem of the shortage of social medical resources,and the application of wearable health monitoring equipment is a powerful measure to solve these problems.Based on the analysis of application scenarios,this thesis designs and implements a set of wearable multi-sensor physiological monitoring system.The system realize the real-time measurement of body temperature,pulse rate,oxygen saturation and blood pressure through the acquisition of ECG signal,temperature signal,pulse wave signal and acceleration signal.The main research contents of this thesis are as follows:1)The overall structure design of the system is completed.The system consists of four interrelated software and hardware modules to realize multi-physiological parameter monitoring functions.Among them: the temperature acquisition module realizes highprecision human skin surface temperature measurement,the pulse wave acquisition module realizes measurement of pulse wave（PPG）and analysis of pulse rate and blood oxygen saturation,the ECG acquisition module collects single-lead ECG signal.The above hardware modules realize the communication and data transmission of the upper computer through Bluetooth,and the ambulatory blood pressure measurement software module based on ECG and PPG is realized in the upper computer.2)Based on the above structure design,the concrete design and implementation of the system hardware module is completed,including sensor selection,circuit design,PCB production and physical structure design,etc.3)To design related drivers,Bluetooth-related service files and low-power implementations for each system hardware module;in addition,the temperature acquisition module has completed the temperature value calculation program design based on the temperature-resistance formula;the pulse wave acquisition module has completed the pulse rate calculation program based on the main pulse wave interval and the blood oxygen saturation calculation program based on the empirical formula are designed;the terminal node part completes the design of the data forwarding program of the Bluetooth Dongle and the host computer program.4)Aiming at the system dynamic blood pressure measurement software module,using the real-time ECG and PPG signals provided by other modules,based on the cuffless characteristic parameter method,using machine learning method to achieve dynamic blood pressure monitoring.First,ECG,PPG and blood pressure data were obtained,and then the original data were preprocessed by wavelet transform.Then 43 characteristic parameters were selected and extracted,and then the regression algorithm contrast experiment was designed to choose the random forest regression method as the way to train the model,and then the hyperparameters are optimized by combining random search and grid search,and finally a random forest model for ambulatory blood pressure prediction was established.5)Perform performance metrics on what the system can achieve.The system can achieve the function of the performance index test.The test results show that the measurement accuracy of each physiological parameter of the system can reach the expected performance index,and the bluetooth transmission performance of the system and the power consumption of each acquisition device in the system can also meet the expected indicators.