Research Of Electrocardiogram And Impedance Cardiography Signals Based On Wavelet Transform

In recent years, with the increasing aging of population, cardiovascular diseases can cause most deaths in all diseases, and they have become one of the main diseases which threaten human’s lives and health. Therefore, early detection and diagnosis of cardiovascular diseases has become the urgent issue in medical field. Because Electrocardiogram (ECG) waveform has regularity and can be detected easily compared with other biomedical signals compared with any other biological signals, therefore, the study and analysis of ECG signal become important security to diagnose diseases. As the development of diversification of heart diseases, and the diseases cannot be diagnosed corretly if physicians only analyze ECG signal, as ECG signal cannot reflex mechanical performance of the heart, so many researchers begin to study Impedance Cardiography (ICG) signal. ICG signal become more and more important because it is noninvasive, convenient to monitor hemodynamic paramters. Comprehensive analysis of ECG and ICG signal help diagnose the patients’ heart diseases accurately according to clinical experiments.The main contents of this paper are the study of ECG and ICG signals by using wavelet transform technology. Wavelet Transform (WT) provides a novel signal processing access, and it has the characteristics of multiresolution. The main contents of this paper are divided into two parts:The first part is the ECG analysis from two aspects of denoising and charicteristics extraction. In terms of denoising, this paper proposed a new threshold function method according to the characteristics of ECG signal, and noises are removed from ECG signal by selecting the appropriate threshold and wavelet function. MIT-BIH arrhythmia database is used for simulation and authentication. Meanwhile, classical algorithms (hard threshold and soft threshold method) were used for denoising ECG signal. Compared with algorithms proposed by other researchers, the novel threshold algorithm has performed significantly better according to simulation results; In terms of characteristics extraction of ECG signal, accurate extraction of P wave, QRS complex and T wave is critical to diagnose the diseases, accurate extraction of the characteristics plays an important role for detection of ECG signal. This paper used wavelet fuction Mexican Hat for characteristics extraction on the fifth scale, and the algorithm was validated by using the MIT-BIH arrhythmia database, extraction accuracy rate is99.78%.The second part is the study of ICG signal, it includes two aspects of denoising and charicteristics extraction. In the respect of denoising ICG signal, considering the ICG signal and ECG signal are micro biomedical signal, so the ICG signal is denoised by same algorithm with ECG signal. This paper used the ECG and ICG monitoing device which was produced by Institute of Electronics of Chinese Academy of Sciences to acquire ICG signal. The novel threshold method, hard thresholding and soft thresholding are used to denoise ICG signal, the novel threshold method has a higher signal-to-noise ratio (SNR) by comparing denoising performance; In the respect of characteristic points extraction, B, C and X points of ICG signal were detected, the algorithm was validated by using denoised ICG signal, extraction accuracy rate is98.7%.