基于盲源提取和希尔伯特变换的心音包络提取

Envelope Extraction of Heart Sound Based on Blind Source Extraction and Hilbert Transform

心音包络比原始心音可以更好地显示心音的特征,是进行心音识别的基础。由于在体表采集心音信号时容易受到呼吸音、外界环境中不确定干扰的影响而且干扰信号不容易被常规滤波器滤除,结果导致提取的心音包络不能真实反映心音的特征。为了准确的提取心音包络,提出了一种新的心音包络提取方法,首先采用盲源提取算法从采集的信号中获取比较纯净的心音信号,然后采用希尔伯特变换提取心音的包络。通过提取合成心音数据和实际采集的心音信号包络证明了上述方法的可靠性。

Heart sound signals（HSS） are interfered easily by respiration sound（RS） and other unknown disturbances during the course of body surface gathering which are hard to be removed by the general filters.So results in envelope extraction of HSS cannot reflect the features of HS.A novel approach was presented in order to obtain exact envelope of heart sound.The approach is based on blind source extraction（BSE） and Hilbert Transform（HT）.Firstly,a clearer HSS was obtained using BSE from gathering signals.Secondly,the envelope of HSS was extracted by Hilbert transform.Reliability of this proposed approach was confirmed by both synthetic signals and real gathering data.