计算机工程与应用 ›› 2015, Vol. 51 ›› Issue (16): 215-218.

• 工程与应用 • 上一篇    下一篇

改进DTW算法的心电信号相似性度量

涂  辉1,刘  丽1,2,张正金1   

  1. 1.江南大学 物联网工程学院,江苏 无锡 214122
    2.无锡市第四人民医院 信息科,江苏 无锡 214122
  • 出版日期:2015-08-15 发布日期:2015-08-14

Improved DTW algorithm to measure similarity of ECG signal

TU Hui1, LIU Li1,2, ZHANG Zhengjin1   

  1. 1.School of IOT Engineering, Jiangnan University, Wuxi, Jiangsu 214122, China
    2.Department of Information, Wuxi Fourth People’s Hospital, Wuxi, Jiangsu 214122, China
  • Online:2015-08-15 Published:2015-08-14

摘要: 针对时间序列相似性度量中动态时间规整(DTW)算法在序列时间轴偏移较大时易产生病态路径及匹配不准确的问题,根据心电信号自身的特性,提出一种基于心电信号最显著特征的优先匹配法。为减小算法时间复杂度,根据心电信号R波特征点位置自适应地划分弯曲路径的约束窗口。与传统算法比较,在提高准确率的同时,运行时间大幅降低。

关键词: 时间序列, 动态时间规整, 心电信号, 优先匹配, 自适应

Abstract: In view of when use the Dynamic Time Warping(DTW) algorithm to measure the time series similarity, but the sequence has a larger excursion it will cause abnormal path and inaccurate matching. According to the nature of the electrocardiogram(ECG) signal, the most significant feature based on ECG priority matching method is proposed. In order to reduce the time consumption, according to the location of R waves, restrictive windows for curved path adaptively divided, compared with the traditional method, the proposed algorithm increases the accuracy, and meanwhile the running time is significantly reduced.

Key words: time series, dynamic time warping, ElectroCardioGram(ECG), priority matching, adaptive