当前位置:首页 > 智慧健康列表 > 科学仪器 > 详细信息
基于非线性理论的ECG 信号多重分形特性谱研究
来源: | 作者: | 发布时间:2014-4-15 17:52:56

基于非线性理论的ECG 信号多重分形特性谱研究
杨小冬
(中国矿业大学计算机科学与技术学院,江苏 徐州 221116)
摘要:生理信号属非平稳时变信号。因此,采用非线性动力学方法能更好的揭示其特性和机理。本文以多重分形方法,详细研究了大量样本同步十二导联心电图(ECG)信号的奇异谱面积和质量指数谱曲率,发现各种人群该特征值有较大差异,代表了不同含义。研究发现,健康年轻人ECG 的奇异谱面积十二导均值最大、心梗患者最小,其他人群处于中等大小水平。表明健康人生理系统处于最佳状态,其ECG 信号非线性动力学复杂程度也最高。随着年龄增长,人体自主神经的自律控制功能逐渐减弱,由多重分形趋向单重分形,人的ECG信号奇异谱面积十二导均值在一定程度上反映了人体自主神经控制的强弱。另外,健康人ECG 信号的质量指数谱曲率(绝对值)最小,CHF 和VF 患者依次增大。本文研究和结论期望为早期诊断和临床应用提供有价值的信息。
关键词:ECG;非线性;多重分形;奇异谱;质量指数谱
中图分类号:R318
Research of ECG multifractal characteristic spectrum based on nonlinear dynamic theorey
YANG Xiaodong
(School of Computer Science and Technology, China University of Mining and Technology,JiangSu XuZhou 221116)
Abstract: Physiological signal belongs to nonstationary and time-variant signal. Thus, Nonlinear dynamic theories can be better to disclose its characteristic and mechanism. In this paper, multifractal approach is used to research the singularity spectrum area and mass exponent spectrum curvature of synchronous 12-lead ECG signals from a largr test set. We find this
eigenvalue is not all the same to the kinds of crowd, and it represents different implications. Among this, the mean value of 12-lead singularity spectrum area from healthy young man is the largest and this value becomes the minimum when it comes to the Myocardial Infarction (MI) sufferer. As to the other crowds, this value is wholly at the middle level. It indicates that the physiological system of healthy young man is at the best status. We also find the self-discipline control of human autonomic nervous system weakens gradually when people get older. It tends to turn to monofractality from multifractality. Thus, the 12-lead mean value of singularity spectrum area of human ECG signal can reflect the self-discipline control status of human autonomic nervous to some degree. In addition, the mass exponent spectrum curvature from healthy man is the least and the value from CHF anf VF sufferer increase accordingly. We expect our research and conclusion can provide some information to early diagnosis and clinical application.
Keywords: ECG; Nonlinear; Multifractality; Singularity spectrum; Mass exponent spectrum