Abstract：In order to solve the problem of end effect in empirical mode decomposition (EMD), this paper proposed method of extracting the features of heart sounds signal through empirical mode decomposition after wavelet denoising. A new method of data extending based on self - adaptive waveform matching was established for EMD. With wavelet denoising, the influence of useless frequency component on the later decomposition was decreased, thus effectively reduced the decomposition layers of the EMD. The self-adaptive waveform matching extending method considered both inner characteristics of signal and variation tendency of endpoint, hence it is more reasonable compared with the conventional extending method. Decompose 40 cases of heart sounds signal with the method, the location of the first and second heart sound was identified by explicit double threshold, the detection rate of S1 was up to 97.05%, and that of S2 was 97.12%. The result indicated that the proposed method could solve the end issue effectively, the speed and accuracy of the EMD were improved.
郭兴明* 袁志会. 基于小波变换和经验模式分解的心音信号研究[J]. 中国生物医学工程学报, 2012, 31(1): 39-44.
GUO Xing Ming* YUAN Zhi Hui. Research on Heart Sounds Signal Based on Wavelet Transform and Empirical Mode Decomposition. journal1, 2012, 31(1): 39-44.
