Multivariable approach to dynamic ECG classification

Marco Gioè*, Dragana Nikolic, Riccardo Caponetto, Branislav Vuksanovic, Maide Bucolo

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The first step towards a heart attack is a condition called Ventricular Arrhythmia. This paper proposes a system that uses the advanced digital signal processing techniques to analyse electrocardiogram (ECG) signals and recognise the Arrhythmia condition. In addition, proposed method can differentiate the more dangerous condition, Ventricular Arrhythmia from a simple Arrhythmia. The proposed technique combines the classical ECG signal parameters (e.g. Heart Rate Variability) with the standard statistical signal parameters, nonlinear parameters used in the fields of Chaos Theory and parameters obtained using Symbolic Analysis techniques. Linear Discriminant Analysis (LDA) is employed in order to reduce the size of ECG parameter set and is followed by a clustering algorithm.

Original languageEnglish
Title of host publicationProceedings of the 18th IFAC World Congress
Pages14301-14306
Number of pages6
Volume44
EditionPART 1
DOIs
Publication statusPublished - 1 Dec 2011
Event18th IFAC World Congress - Milano, Italy
Duration: 28 Aug 20112 Sep 2011

Conference

Conference18th IFAC World Congress
CountryItaly
CityMilano
Period28/08/112/09/11

Keywords

  • Classification
  • Clustering
  • Nonlinear analysis
  • Symbolic analysis

Fingerprint

Dive into the research topics of 'Multivariable approach to dynamic ECG classification'. Together they form a unique fingerprint.

Cite this