Analysis Of Abnormal ECG Signal Caused By Abnormal Cardiac Pacing Area

With the rapid development of modern society,the speed of life changed rapidly,the pressure of heart load is increasing,and some bad living habits make the incidence of cardiovascular disease high,so heart disease have become one of the most seriously diseases that endanger people’s health.In our country,approximately 20% to30% of deaths are related to heart disease.Estimating of human heart disease is benefitting great significance for timely treatment and prevention of sudden death from heart disease.At present,the diagnosis of cardiovascular diseases basically depend on conventional electrocardiogram(ECG)as the main method,so accurate and rapid interpretation of the electrocardiogram is very important.However,the current situation in our country is that medical resources are still extremely insufficient.The number of doctors per capita is far below the global average,and there is a serious imbalance in medical resources and levels.Large hospitals have advanced equipment,high levels of diagnosis and treatment,and basic hospitals are backward in equipment,diagnosis and treatment.Therefore,misdiagnosis and missed diagnosis are prone to occur.With the development of science and technology,Various automatic recognition and auxiliary diagnosis systems are invited and greatly alleviate and improve this situation.This paper mainly focuses on the intelligent analysis of abnormal ECG signals:(1)Design the ECG signal preprocessing and algorithm.This article has fully studied the classification and characteristics of noise in ECG signals,and designed denoising schemes to eliminate EMG interference,power frequency interference and baseline drift noise.The simulation experiment verifies the validity of the ECG signal preprocessing algorithm designed in this thesis.(2)Design the ECG signal waveform detection algorithm.In this thesis,the wavelet decomposition method is used to decompose the ECG signal in multiple layers.According to the different wavelet characteristics of different waveforms,the peak position is checked at different levels.Based on the location of the peak position,combined with the average ECG cycle,the waveform is searched for Start and end positions.The simulation experiment verifies the efficiency and robustness of the ECG waveform detection algorithm.(3)Research the classification algorithm of abnormal ECG signals.Here we use five kinds of ECG signal characteristics,combined with the support vector machine method to complete the classification of the ECG signal,and discusses the performance effects of the four kinds of ECG signals(N,S,V,F).Further research on single feature and multiple feature combinations,combined with different combination principles,vector machine for classification and recognition accuracy.The simulation results show that the classification and recognition effects of vector machines under different conditions are different.In general,the more features,the more accurate the classification and recognition results.