【摘要】 提出了一种新的时间序列转换成网络的方法,从而可以通过网络拓扑统特征来刻画时间序列。将该方法应用于心室纤颤的检测中,发现健康个体与室颤病人相应网络非常明显的不同,将网络的聚类系数与同配系数作为联合指标可以区分两者。此外,还将该算法应用于HRV分析,结果表明,算法能够区分不同病态下的心跳间隔信号。更多还原

【Abstract】 This paper proposes a map from time series to a network,making it possible to use network statistics to characterize time series. With applying our method to ventricular fibrillation detection,we find the clustering coefficient and the assortative coefficient of associated networks could distinguish between healthy subjects and ventricular fibrillation patients. Besides,this paper notes that our algorithm could distinguish healthy group,congestive heart failure patients and atrial fibrillation patients.更多还原