Time-frequency analysis of heart sounds in telemedicine consulting system for auscultation

With the technological improvement of modern communication, telemedicine has obtained fast development. Telemedicine provides an opportunity for patients using computers and Internet to connect with doctors who may be thousands of miles away. The heart sound signal, which is one of the most important physiological signals of human beings, reflects the conditions of human heart, and it has more advantages than ECG in some ways. Nowadays, computer-aided diagnosis which depends on digital signal processing becomes more and more popular; the computer-aided diagnosis of heart sound becomes available now. In this research, remote auscultation and computer-aided diagnosis of heart sound are combined together. This paper deals with heart sounds based on Hilbert-Huang transform from analysis of time domain and frequency domain, in order to extract a series parameters which are useful for Computer-Aided diagnosis. Firstly, heart sounds are preprocessed by using the wavelet transform and Huang transform technology. Wavelet threshold denoising can effectively remove the noise of heart sounds. Huang transform can extract a series of intrinsic mode functions, and choose appropriate intrinsic modal functions, which can effectively remove the low-frequency noise of signal. This paper also studies the extraction of envelopes of the heart sounds based on Hilbert transform. The heart sounds are segmented effectively based on Hilbert envelops, thus time domain features of the heart sounds can be more accurate extracted. Welch power spectrum ratio of high frequency heart sounds and low frequency heart sounds can be extracted, so that Computer-Aided diagnosis can be designed and achieved.