Detection of transient episodes in heart rate variability signals

Author

Obayya, Marwa ; Abou-Chadi, Fatma

Author_Institution

Dept. of Electron. & Commun. Eng., Mansoura Univ., Egypt

fYear

2004

fDate

16-18 March 2004

Lastpage

42377

Abstract

This paper compares the performance of four approaches for the detection of transient episodes in the heart rate variability (HRV) records. These are based on autoregressive (AR) modeling, discrete wavelet transforms (DWT), wavelet packet transforms, and hidden Markov modeling (HMM). A competitive neural network has been applied for classification and the results of the four techniques have been compared. It has been concluded that the autoregressive model is the most efficient technique for detecting the essential features describing the transient episodes in HRV.

Keywords

autoregressive processes; discrete wavelet transforms; electrocardiography; feature extraction; hidden Markov models; neural nets; signal detection; unsupervised learning; DWT; HMM; HRV record; autoregressive modeling; competitive neural network; discrete wavelet transform; heart rate variability signal; hidden Markov modeling; transient episode detection; wavelet packet transform; Biological neural networks; Biological system modeling; Discrete wavelet transforms; Heart rate; Heart rate detection; Heart rate variability; Hidden Markov models; Rhythm; Testing; Wavelet packets;

fLanguage

English

Publisher

ieee

Conference_Titel

Radio Science Conference, 2004. NRSC 2004. Proceedings of the Twenty-First National

Print_ISBN

977-5031-77-X

Type

conf

DOI

10.1109/NRSC.2004.1321873

Filename

1321873