北京生物医学工程

病理心电信号的Jensen-Shannon_divergence分析

Complexity analysis of pathological ECG signal based on Jensen-Shannon divergence

作者：许欢孟浩王俊

单位：南京邮电大学地理与生物信息学院图像处理与图像通信重点实验室(南京210003)

关键词： Jensen-Shannon divergence；复杂度；房性早搏；窦性心动过缓；电信号

分类号：

出版年·卷·期（页码）：2012·31·6（565-569）

摘要：

研究快速区分不同病理心电信号的算法是当前临床医学诊断的研究热点之一。方法本文采用基于Jensen-Shannon divergence的复杂度分析方法，从MIT-BIH标准数据库中提取正常窦性心率、房性早搏、窦性心动过缓信号，分别计算它们的复杂度。结果正常心率信号复杂度最高，窦性心动过缓信号次之，房性早搏最低。方差检验结果表明，此方法得出的3种信号的复杂度具有显著性差异。结论病理心电信号的Jensen-Shannon divergence分析对临床医学检测和诊断房性早搏、窦性心动过缓信号有很好的借鉴意义。

Objective Algorithm research on high-speed diagnosis of pathological ECG is one of the current research hotspots in clinical diagnosis. Methods In this paper,complexity measure based on Jensen-Shannon divergence was used to compute complexity of the ECG signals,which include normal sinus rhythm,atrial premature contraction (APC) and sinus bradycardia (SBR) signals from the MIT-BIH standard database. Results The results showed that the three kinds of signals had different complexity measures. Normal sinus rhythm had the highest complexity,followed by SBR signals and APC signals. The variance test indicated that this analysis could disclose the significant differences among these three signals’ complexity. Conclusions This complexity analysis based on Jensen-Shannon divergence is a good reference for clinical detection and diagnosis of APC and SBR signals.

参考文献：

［1］Lin J. Divergence measures based on the Shannon entropy［J］. IEEE transaction on information theroy,1991,37 (1):145-151.
［2］Lamberti PW,Majtey AP. Non-logarithmic Jensen–Shannon divergence［J］. Physica A,2003,329(1-2):81-90.
［3］Hung WL,Yang MS. On the J-divergence of intuitionistic fuzzy sets with its application to pattern recognition［J］. Information Sciences,2008,178 (6):1641-1650.
［4］张景祥,徐振源,赵晓兵. β- KL离散量在识别DNA编码区域中的应用［J］. 生物信息学报,2010,8(2):110-113.
Zhang Jingxiang,Xu Zhenyuan,Zhao Xiaobin. Study on the methods for finding borders between coding and noncoding DNA regions［J］. China Journal of Bioinformatics,2010,8(2):110-113.
［5］孙桂玲. 稳心颗粒治疗窦性心动过缓合并室性早搏的临床观察［J］. 中国现代药物应用,2012,6(11):21-22.
Sun Guiling. Wenxinkeli treatment of sinus bradycardia associated with the dinical observation of premature ventricular contractions［J］. China Journal of Modern Drug Application,2012,6(11):21-22.
［6］边越. 动态心电图观察阵发性心房颤动的发生机制［J］. 现代电生理学杂志,2011,18(2):91-93.
Bian Yue. The observation of mechanism of paroxysmal atrial fibrillation by dynamic electrocardiogram［J］. Journal of Modern Electrophysiology,2011,18(2):91-93.
［7］孙荣荣,汪源源,方祖祥. 基于灰色关联度K-近邻法的房性心律失常识别［J］. 航天医学与医学工程,2007,20(3):193-197.
Sun Rongrong,Wang Yuanyuan,Fang Zuxiang. Atrial arrhythmia recognition based on grey correlation and K-nearest neighbour［J］. Space Medicine & Medical Engineering,2007,20(3):193-197.
［8］Physiobank archive index［EB/OL］. ［2012-08-12］. http://www. physionet. org/physiobank/database.
［9］MartinMT,Plastino A,Rosso OA. Generalized statistical complexity measures:geometrical and analytical properties［J］. Physica A,2006,369:439-462.
［10］Rosso OA,Larrondo HA,Martin MT,et al. Distinguishing noise from chaos［J］. Phyisical Review Letters,2007,99,154102.
［11］戚建新,卞正中,杨强. 基于微机的心电信号实时自动分析系统［J］. 北京生物医学工程,1997,16(3):157-160.
Qi Jianxin,Bian Zhengzhong,Yang Qiang. A microcomputer-based realtime automatic analysis system for ECG signal［J］. Beijing Biomedical Engineering,1997,16(3):157-160.
［12］胡鹏,张永红,张菊鹏,等. 基于小波变换的心律失常判别算法［J］. 北京生物医学工程,2003,22(1):23-26.
Hu Peng,Zhang Yonghong,Zhang Jupeng,et al. A wavelet-based arrhythmia detection method［J］. Beijing Biomedical Engineering,2003,22(1):23-26.
［13］Li J,Ning XB. Dynamical complexity detection in short-term physiological series using base-scale entropy［J］. Physical Review E,2006,73:052902.
［14］Plastino A,Lamberti PW,Martin MT,et al. Intensive entropic non-triviality measure［J］. Physica A,2004,334 (1-2):119-131.

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录