Analysis of Diastolic Murmurs for Coronary Artery Diseasebased on Hilbert Huang Transform

Coronary artery disease (CAD) is the leading disease of heart attacks. A novel approach based on Hilbert-Huang transform (HHT) is proposed to analyze diastolic murmurs of CAD. HHT is composed of empirical mode decomposition (EMD) and Hilbert transform. It is a powerful method for the analysis of nonlinear and non-stationary signal. EMD method is very sensitive to end conditions. Firstly a new improved strategy is proposed to restrict the end effect; then the diastolic murmurs of normal object and CAD patient are analyzed by improved Hilbert Huang transform. Hilbert spectrum and marginal spectrum are studied. The results show that the Hilbert spectrum and marginal spectrum reveal not only the time-frequency varying characteristic of diastolic murmurs but also more physically meaningful interpretations of the underlying hemodynamic processes.