A Study And Application Of Using Wavelet Transform In The Processing Of Electrocardiogram Signals

The paper mainly studies the application of using wavelet transform in the processing of electrocardiogram signals. The wavelet transform is the latest time-domain techniques that priors to traditional methods. Wavelet transform allows a signal to be decomposed such that frequency characteristics and the location of particular features in a time series may be highlighted simultaneously. So any particular part of the signal can be located for analyzing. This made the wavelet transform to be widely used in the signal processing area. Applying the wavelet transform in medical signals is the breakthrough in the tools and techniques.This paper adopts the viewpoint of the signal processing to present the continuously wavelet transform, discrete wavelet transform and multi-resolution analysis. The following is our research and experiment of using wavelet transform in the processing of electrocardiogram signals.(1) Noise cancellationThe electrocardiogram signal often has the following three forms of noise: 60Hz power line interface, muscle noise and baseline wander. By using the wavelet transform the three main interferences can be removed. But the filtering method may affect the QRS complex. A new noise cancellation algorithm is proposed in the paper. Our experimental results demonstrate that this is an efficient filtering method.(2) Electrocardiogram data compressionThe need for compressing biomedical signal is important due to the tremendous amount of data that need to be stored efficiently at low cost. In this paper two kinds of electrocardiogram data compression scenarios are presented. The one is lossless data compression method and the other is semi-lossless data compression algorithm based on wavelet transform. For the latter scenarios, the quality of reconstructed electrocardiogram signal is nearly the same of the original one.(3) Extracting the baby electrocardiogramThe Electrocardiogram confused of mother and baby is studied. The baby electrocardiogram that has been extracted is filtered by lossless QRS complex method that proposed in this paper. The result shows that the interference of mother has been removed in the largerest extent and the clear baby electrocardiogram signal has been acquired.(4) The detection of abnormal intra-QRS complexThe paper introduces a method of extracting high frequency components of abnormal intra- QRS complex by using wavelet transform. The difference with the high-frequency part of the QRS complex is analyzed as quantitative index.In the end of paper the Electrocardiogram Data Processing Software System is introduced. This software can be used as a researching platform to test algorithms related to electrocardiogram signals and can complete all of the algorithms that this paper proposed.