基于Hilbert-Huang Transform的心音信号谱分析

心音信号是一种典型的非平稳信号,传统信号处理方法的应用受到很大限制.针对此本文提出了基于Hilbert-Huang Transform(HHT) 的心音信号的分析方法,对冠心病患者的心音信号进行了分析.通过把心音信号分解为内蕴模式函数,利用Hilbert变换建立了心音信号的时间-频率-能量三维Hilbert谱分布以及边界谱分布;Hilbert谱及其边界谱在时域以及频域以较高的分辨率表征了心音信号的时频变化特性,揭示了冠心病患者心音信号的病理特征;为冠心病的早期无损诊断奠定了坚实基础,临床实践中有较大的指导价值.

Spectrum Analysis of Heart Sound Signal Based on Hilbert-Huang Transform

Heart sound signal is typical non-stationary signal; the traditional signal processing methods are constrained. A novel method based on Hilbert-Huang Transform(HHT) is introduced. The heart sound signal of coronary artery disease is analyzed. First, decomposing the heart sound signal into a little number of intrinsic mode function (IMF), second performing the Hilbert transform an each IMF component; then achieving the three-dimensional Hilbert spectrum and marginal spectrum, where both the amplitude and frequency are functions of time. The Hilbert spectrum and marginal spectrum provide high-resolution time-varying characteristic of heart sound signal, the result shows the clinical characteristic of CAD. It establishes the basis for the early diagnosis of heart diseases.