Nonlinearity parameters for the classification of high risk Myocardial Infarction subjects

Maria G. Signorini, Federico Lombardi, Sergio Cerutti

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

The paper presents the analysis of the Heart Rate Variability (HRV) signal in 19 subjects who recently had a Myocardial Infarction episode (MI). The study follows a nonlinear approach based on the multiparametric analysis of some invariant properties of the dynamical system generating the time series. First we reconstruct the system embedding space from the HRV time series. The False Nearest Neighbors (FNN) criterion provides the real embedding dimension value. Results show that through the FNN method it is possible to identify the correct number of LE in the system. Parameter values significantly separate subjects who after MI keep a good performance of the cardiac pump (normal ventricular ejection function, NEF) vs. the group which after MI shows a reduced ventricular ejection fraction (REF).

Original languageEnglish
Title of host publicationComputers in Cardiology
Pages545-548
Number of pages4
Volume0
Edition0
Publication statusPublished - 1998

ASJC Scopus subject areas

  • Software
  • Cardiology and Cardiovascular Medicine

Fingerprint Dive into the research topics of 'Nonlinearity parameters for the classification of high risk Myocardial Infarction subjects'. Together they form a unique fingerprint.

  • Cite this

    Signorini, M. G., Lombardi, F., & Cerutti, S. (1998). Nonlinearity parameters for the classification of high risk Myocardial Infarction subjects. In Computers in Cardiology (0 ed., Vol. 0, pp. 545-548)