Classification of Arrhythmia using Multi-Class Support Vector Machine

Tonghui Li, Jieming Ma, Xinyu Pan, Yujia Zhai, Ka Lok Man

Research output: Chapter in Book or Report/Conference proceedingConference Proceedingpeer-review

6 Citations (Scopus)


Arrhythmia has become the most common disease in the medical field. Manual diagnosis of arrhythmia beats is very tedious owing to its nonlinear and complex nature of electrocardiogram (ECG). In this article, a multi-class support vector machine (MSVM) based approach is proposed to solve ECG multi-classification problem. Based on the characteristics of the R-R interval, it has the capability of detecting normal heart rate (NOR), left bundle branch block (LBBB), right bundle branch block (RBBB), atrial premature complex (APC) and Ventricular premature beat (VPC) was mainly discussed. Using ECG MIT-BIH database, simulation results show the proposed method achieves a very high classification accuracy.

Original languageEnglish
Title of host publicationProceedings of the International MultiConference of Engineers and Computer Scientists 2017, IMECS 2017
EditorsOscar Castillo, S. I. Ao, Craig Douglas, David Dagan Feng, A. M. Korsunsky
PublisherNewswood Limited
Number of pages4
ISBN (Electronic)9789881404770
Publication statusPublished - 2017
Event2017 International MultiConference of Engineers and Computer Scientists, IMECS 2017 - Hong Kong, Hong Kong
Duration: 15 Mar 201717 Mar 2017

Publication series

NameLecture Notes in Engineering and Computer Science
ISSN (Print)2078-0958


Conference2017 International MultiConference of Engineers and Computer Scientists, IMECS 2017
Country/TerritoryHong Kong
CityHong Kong


  • Arrhythmia
  • Classification
  • ECG
  • Kernel function
  • MSVM

Cite this