Automatic Classification of Heartbeats Using ECG Signals via Higher Order Hidden Markov Model

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

1 Scopus citations

Abstract

Analysis of heartbeat signals is critical to provide information for the diagnosis of cardiac diseases, e.g., beat detection, segmentation and classification based on electrocardiogram (ECG) signals. This paper proposes an automatic heartbeat classification system based on higher order hidden Markov models (HOHMMs). The proposed system consists of four stages: ECG signal preprocessing stage, HOHMM learning stage, decoding stage, and classification stage. The HOHMM extends the basic hidden Markov model (HMM) by allowing the hidden state to depend on its more distant past, which is used to identify heartbeat patterns utilizing the collected ECG data. The learned HOHMMs are then used to decode and classify new heartbeats with unknown types. A case study is conducted to evaluate the classification performance of the proposed system using MIT-BIH Arrhythmia database. Experimental results show that the developed classification system performs reasonably well, especially for arrhythmia detection.

Original languageEnglish
Title of host publication2020 IEEE 16th International Conference on Automation Science and Engineering, CASE 2020
PublisherIEEE Computer Society
Pages69-74
Number of pages6
ISBN (Electronic)9781728169040
DOIs
StatePublished - Aug 2020
Event16th IEEE International Conference on Automation Science and Engineering, CASE 2020 - Hong Kong, Hong Kong
Duration: Aug 20 2020Aug 21 2020

Publication series

NameIEEE International Conference on Automation Science and Engineering
Volume2020-August
ISSN (Print)2161-8070
ISSN (Electronic)2161-8089

Conference

Conference16th IEEE International Conference on Automation Science and Engineering, CASE 2020
Country/TerritoryHong Kong
CityHong Kong
Period08/20/2008/21/20

Keywords

  • ECG
  • HOHMM
  • MITBIH Arrhythmia database
  • heartbeat classification

Fingerprint

Dive into the research topics of 'Automatic Classification of Heartbeats Using ECG Signals via Higher Order Hidden Markov Model'. Together they form a unique fingerprint.

Cite this