非线性心音时间序列的最大Lyapunov指数

Largest Lyapunov Exponents of Nonlinear Heartbeat Time Series

许多关于心脏运动的研究表明,心脏运动中的确存在着混沌现象。以非线性时间序列最大Lyapunov指数的计算为主线,分别使用相空间重构技术、C-C方法的延迟时间和嵌入维数估计、Wolf法最大Lyapunov指数计算,提出了用均值优化最大Lyapunov指数的方法,并针对正常心音、二尖瓣不全、心音分裂和主动脉收缩异常等4种心音信号的最大Lyapunov指数进行计算和分析。结果表明,这些心音信号中都存在混沌现象。其中,正常心音信号最大Lyapunov指数值最大,其他3个较小,说明正常的心脏运动的混沌程度要比异常时的混沌程度高。

Research has shown existence of chaos in heart movements.In this paper,we discuss the calculation of the largest Lyapunov exponents（LLE） of the nonlinear time series,and use the C-C method to estimate time delay and embedding dimension to reconstruct the phase spaces.We use the Wolf method to calculate the LLE.We present a way to optimize the exponents using a mean value method.Different heartbeat signals are studied,and their LLE calculated.The signals analyzed including healthy heartbeats,split of second heart sound,mitral incompetence,and abnormal aortic shrinkage.The results reveal nonlinearity and chaotic characteristics in the dynamic heart movement.The LLE results show that healthy heart has more obvious chaotic movements than abnormal hearts.