基于小波多分辨分析的第一第二心音提取

在心音信号的分析中,为了有针对性的分析第一心音(S1)、第二心音(S2)、第三心音(S3)、第四心音(S4),首先需要将它们从采集的心音信号中分离出来.本文提出一种方法,利用小波多分辨分析,提取第一心音及第二心音的同步信号,从而完成对第一心音和第二心音的实时分离.该方法区别于传统的依靠心电信号进行同步提取第一心音和第二心音的方法,避免了采集心音信号的同时需采集心电信号的麻烦,也提供了一种实现信号自同步的思路.用本方法对28例心音信号进行了仿真实验,1、S2均能被正确地分离出,表明该方法是可行的.

Extraction of the First and the Second Heart Sounds Based on Multi-resolution Analysis of Wavelet Transform

In order to analyze the first,the second,the third,or the forth heart sounds (s1,s2,s3,s4),it is necessary to separate them out from the phonocardiogram(PCG) respectively.In this paper is introduced a method,by which s1 and s2 can be separated out from PCG,using multi-resolution analysis of the wavelet transform to extract the synchronous signals of the s1 and s2.The separation of s1 and s2 is done in real time.The method is different from the traditional synchronization based on electrocardiogram(ECG) to extract s1 and s2.Instead of sampling both PCG and ECG simultaneously,it is just necessary to sample PCG only.It also gives us an inspiration that a signal itself can be used as a synchronous signal.This method has been used to separate s1 and s2 for 28 PCG cases successfully(simulation experiment).