Research And Application Of ECG Classification Based On BP Neural Network

With the change of people's lifestyle,the aging population is becoming more and more serious,and the incidence of cardiovascular diseases is gradually increasing.Effective and accurate prevention and cure of cardiovascular diseases has become an important problem to be solved urgently.Mobile Electrocardiograph(ECG)monitoring can be used for long-term monitoring of patients,and it has a positive significance for the prevention and treatment of cardiovascular diseases.The key problem of ECG monitoring is whether the ECG signal can be classified effectively and accurately.ECG signal classification algorithm can be used to process and analysis the ECG signal,complete the process of automatic diagnosis of heart disease,plays a major role in the prevention of cardiovascular disease,the design and implementation of more accurate and effective ECG automatic classification algorithm has very important research significance.Therefore,this paper proposes an automatic recognition and classification algorithm of ECG signals based on BP neural network.Based on the algorithm,we carry out the research of mobile ECG monitoring and early warning system,and design a mobile ECG monitoring early warning system based on BP neural network.This paper first describes the background and significance of the classification of ECG signals and the current research status of ECG signal detection and classification early warning algorithms.And then introduces the basic principle of BP neural network and BP neural network in mind to the present situation of the application of electrical signal recognition classification.An automatic recognition and classification algorithm for atrial premature beat and ventricular premature beat based on BP neural network is proposed in this paper.The algorithm uses MIT-BIH international standard ECG data as a data source.First of all,the Pan&Tompkins algorithm is used to detect ECG signals.After the preprocessing of the feature extraction,the character parameters,and then input to the BP neural network classification model with two hidden layer is designed in this paper,experiment.The experimental results show that the algorithm can achieve 93.33% accuracy for automatic recognition and classification of atrial premature and ventricular premature,which has reference significance and can be applied to practice.Finally,this paper presents a design of mobile ECG monitoring early warning system,completed the front-end hardware acquisition module,network communication module,the design and implementation of mobile client software,the designed functions can be realized,to provide the reference for the study of other ECG signal classification and its application.