Parametric time-domain measures of long-term heart rate variability signal

Author

Saliu, Sokol ; Poci, Dritan ; Edvardsson, Nils

Author_Institution

Chalmers Univ. of Technol., Lindholmen Coll., Goteborg, Sweden

Volume

2

fYear

2001

fDate

2001

Firstpage

1929

Abstract

Based on alpha-stable distributional modelling, we have proposed a set of parametric measures to analyze long-term heart rate variability(HRV). The set, in difference from standard measures, accommodates premature beats in the HRV analysis. It preserves the structure of standard measures, redefines many of them and includes new measures. The rationale of the method was demonstrated by simulations. We found that the standard measures SDANN and SDNNIndex were strongly correlated to their parametric counterparts SDARR and SRRIndex, DRRIndex in a group of normal subjects (correlation coefficients 0.979, 0.939, 0.999) and less so in a group of patients (0.942, 0.544, 0.463). Statistical tests indicated that the new parametric measures could better differentiate better between the two groups.

Keywords

electrocardiography; medical signal processing; probability; statistical analysis; time-domain analysis; Monte Carlo simulations; alpha-stable distributional modelling; generalized central limit theorem; heavy-tailed distributions; long-term heart rate variability signal; parametric time-domain measures; premature beats; statistical measures; whitening; Cardiology; Educational institutions; Heart rate; Heart rate variability; Hospitals; Measurement standards; Neural networks; Signal analysis; Testing; Time domain analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society, 2001. Proceedings of the 23rd Annual International Conference of the IEEE

ISSN

1094-687X

Print_ISBN

0-7803-7211-5

Type

conf

DOI

10.1109/IEMBS.2001.1020604

Filename

1020604