期刊名称：Journal of Automation, Mobile Robotics & Intelligent Systems (JAMRIS)
印刷版ISSN：1897-8649
电子版ISSN：2080-2145
出版年度：2018
卷号：12
期号：1
页码：23
出版社：Industrial Research Inst. for Automation and Measurements, Warsaw
摘要：Analysis of electrocardiogram and heart rate providesuseful information about health condition of a patient.The North Sea Bicycle Race is an annual cycling competitionin Norway. Examination of ECG recordings collectedfrom participants of this race may allow defining andevaluating the relationship between physical enduranceexercises and heart electrophysiology. Parameters reflectingpotentially alarming deviations are to be identified inthis study. This paper presents results of a time-domainanalysis of ECG data collected in 2014, implementingK-Means clustering. A double stage analysis strategy,aimed at producing hierarchical clusters, is proposed. Thefirst phase allows rough separation of data. Second stageis applied to reveal internal structure of the majorityclusters. In both steps, discrepancies driving the separationcould stem from three sources. Firstly, they could besigns of abnormalities in electrical activity of the heart.Secondly, they may allow discriminating between naturalgroups of participants – according to sex, age, physicalfitness. Finally, some deviations could result from faultsin data extraction, therefore serving in evaluation of theparameters. The clusters were defined predominantly bycombinations of features: heartbeat signals correlation,P-wave shape, and RR intervals; none of the featuresalone was discriminative for all the clusters.
关键词：ECG; principal Component Analysis; silhouette;analysis; clustering
