基于小波变换的心音包络提取算法及应用

背景：心音信号包含了大量心脏瓣膜活动的生理信息,心音分析对诊断心脏疾病具有重要的临床意义。目的：旨在通过心音的包络提取,分析心音信号的各种特征,进而判断心音中是否包含杂音,以改善传统听诊技术高度依赖医生经验、听诊范围受限的缺点。方法：提出了一种采用小波变换来提取心音包络的方法,通过与采用希尔伯特-黄变换、数学形态学、平均香农能量等心音包络求解方法进行对比,证明这种方法具有算法简便、曲线光滑、特征点突出等优点。结果与结论：将该方法用于临床真实心音的包络提取,利用支持向量机来训练所提取心音包络的面积和小波能量两个特征参数,判别心音信号是否明显包含杂音。选用35例心音数据对算法进行验证,结果表明该算法的准确率达到95%,具有很强的实用性。

Envelope extraction algorithm and phonocardiogram signal application based on wavelet transform

BACKGROUND：The activity of heart valves can be reflected by cardiac sounds,even some heart disease can also find expression in the abnormal heart sounds.So heart sounds analysis has important clinical significance.OBJECTIVE：Through extraction envelope and analysis of the various features of heart sounds,to detect whether there is noise or not in phonocardiogram signals so as to improve the weakness of traditional auscultation technology such as high dependence on the doctors’ experience and the limited auscultation range.METHODS：Extraction envelope curve is one of the commonly used methods to analyze heart sounds.A new method based on wavelet transform to extract the heart sound signals envelope was presented,in contrast to the common methods as Hilbert-Huang transform（HHT） ,mathematical morphology and average Shannon energy.Through practice,the method was proved to contain many advantages：simple algorithm,smooth feature,outstanding feature point.RESULTS AND CONCLUSION：In order to test the accuracy of discriminating normal and abnormal heart sounds,35 heart sounds were collected and analyzed.The experiment demonstrated that the accuracy performances were achieved by 95%,which is very useful in many aspects.