以統計特性進行心跳序列之叢集處理

Clustering Heart Beats Using Heart Rate Variability

黃于珍(Yu-Chen Huang)

指導教授 ： 林志麟

元智大學/資訊學院/資訊管理學系/碩士(2008年)

摘要

本研究提出使用心率變異度(Heart Rate Variability, HRV)的八個統計值(平均值、SDNN、SDNN Index、SDANN、RMSSD、CV、pNN50
與pNN20)來計算兩RR Interval 時間序列間的距離，進而根據此種距
離計算的結果進行RR Interval 時間序列的叢集處理。透過多組不同類型的RR Interval時間序列資料集的實驗結果顯示，此種方式所產生的叢集結果能有效地區隔出資料集內不同類型的RR Interval時間序列。

關鍵字

心率變異度；叢集分析

並列摘要

This study proposes a method to cluster RR Interval time series. First,the distance of any two RR Interval time series is defined based on eight statistic features of Heart Rate Variability (including Mean, SDNN, SDNN
Index, SDANN, RMSSD, CV, pNN50 and pNN20).
Then, RR Interval time series are clustered based on this distance. Experimental results with various datasets indicate that this method can effectively distinguish different types of RR Interval time series.

並列關鍵字

Heart Rate Variability；Cluster Analysis

參考文獻

[1] Airaksinen K.E., Ikäheimo M.J., Linnaluoto M.K., Niemelä M., andTakkunen J.T., “Impaired vagal heart rate control in coronary arterydisease,” Br Heart J,58(6), 592–597, December, 1987.

Goldberger, and C.-K. Peng, “Linguistic Analysis of the HumanHeartbeat Using Frequency and Rank Order Statistics,” PhysicalReview Letters, 90, 108103, 2003.

[3] C.-K Peng, Albert C.-C. Yang, Ary L. Goldberger, “Statistical physicsapproach to categorize biologic signals: From heart rate dynamics toDNA sequences,” CHAOS, 17, 015115, 2007.

[4] Congestive Heart Failure RR Interval Database.

[7] Feinberg WM, Blackshear JL, Laupacis A, et al., “Prevalence, age distribution, and gender of patients with atrial fibrillation: analysis and implication,” Arch Intern Med, 155,469-473, 1995.