Improved methods for detecting main components of heart sounds

The heart is one of the key organs of human bodies and each component of heart sounds reflects important information about the cardiac status. In this paper, an improved method based on envelope for detecting them without reference signals was proposed. First of all, a novel de-noising method using thresholding function in wavelet domain was used to remove background noise and white noise. Secondly, original envelope was got according to Hilbert Transfer and then the final smooth envelope was extracted using Cubic polynomial interpolation. Thirdly, some strategies were taken to get the correct locations of main components of heart sounds in time domain, especially a new method based on window period for recovering lost peaks was discussed. The algorithm was tested by 223 records and the correct rate was 94.6%.