Design And Research Of ECG Signal Automatic Recognition Algorithm And ECG Monitoring System

In recent years,people’s fast-paced and unhealthy lifestyle makes cardiovascular diseases more and more popular,which seriously endangers the health and life of all mankind.Electrocardiogram(ECG)diagnostic technique is widely used to detect and diagnose various cardiac diseases,such as arrhythmias.With the progress and development of electronic science and information technology,the automatic recognition technology of ECG signal has been widely used in the research of heart disease monitoring and diagnosis.In this paper,the automatic recognition algorithm of ECG signal is studied deeply,and the software and hardware of ECG monitoring system is designed.The main research work of this paper is as follows:(1)Aiming at all kinds of interference and noise in the process of ECG signal acquisition,the pretreatment method and signal quality assessment method of ECG signal are designed.The preprocessing method filters the signal sufficiently according to the characteristics of each type of noise.The signal quality evaluation method gives the evaluation index of signal pollution degree by noise.(2)A QRS detection algorithm based on decision tree and check was proposed to solve the problems of poor anti-interference,high rate of missed detection and false detection in the traditional ECG QRS detection method.The algorithm designs a series of check rules such as peak validity detection,cooling window and adaptive threshold to reduce the possibility of missed detection and false detection.The test results show that the QRS detection algorithm in this paper has higher detection sensitivity and accuracy,and greatly reduces the missed and false detection of R wave.(3)Aiming at the problems of difficult feature extraction,low accuracy of automatic classification model and poor practical application caused by the complex morphology of ECG signals,a recognition and classification method based on U-NET full convolutional neural network was designed and simulated.This method does not require the design of artificial features.Through the encoding operation rules of full convolutional neural network,the ECG signal slice data is taken as input and the label map as output,and the position and category of cardiac beats in the signal fragments can be classified.The simulation results show that the ECG signal recognition method in this paper achieves high accuracy in the five classification problems of arrhythmia,and realizes the effective recognition of arrhythmia beats.(4)The design scheme of ECG signal monitoring system is presented,the hardware layout is optimized to reduce the volume and power consumption of the acquisition terminal,and the corresponding software is designed.By designing ECG signal acquisition circuit,wireless network transmission circuit and ECG signal monitoring software,the system structure of "Acquisition terminal-User control terminal-Monitoring terminal" is built,which realizes the functions of ECG signal acquisition,transmission,automatic identification and information display.