A novel method for the detection of apnea and hypopnea events in respiration signals

Péter Várady, Tamás Micsik, Sándor Benedek, Z. Benyó

Research output: Contribution to journalArticle

129 Citations (Scopus)

Abstract

The monitoring of breathing dynamics is an essential diagnostic tool in various clinical environments, such as sleep diagnostics, intensive care and neonatal monitoring. This paper introduces an innovative signal classification method that is capable of on-line detection of the presence or absence of normal breathing. Four different artificial neural networks are presented for the recognition of three different patterns in the respiration signals (normal breathing, hypopnea, and apnea). Two networks process the normalized respiration signals directly, while another two use sophisticatedly preprocessed signals. The development of the networks was based on training sets from the polysomnographic records of nine different patients. The detection performance of the networks was tested and compared by using up to 8000 untrained breathing patterns from 16 different patients. The networks which classified the preprocessed respiration signals produced an average detection performance of over 90%. In the light of the moderate computational power used, the presented method is not only viable in clinical polysomnographs and respiration monitors, but also in portable devices.

Original languageEnglish
Pages (from-to)936-942
Number of pages7
JournalIEEE Transactions on Biomedical Engineering
Volume49
Issue number9
DOIs
Publication statusPublished - Sep 2002

Fingerprint

Neonatal monitoring
Neural networks
Monitoring
Sleep

Keywords

  • Classification
  • Neural networks
  • Polysomnography
  • Respiration monitoring
  • Sleep apnea

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

A novel method for the detection of apnea and hypopnea events in respiration signals. / Várady, Péter; Micsik, Tamás; Benedek, Sándor; Benyó, Z.

In: IEEE Transactions on Biomedical Engineering, Vol. 49, No. 9, 09.2002, p. 936-942.

Research output: Contribution to journalArticle

Várady, Péter ; Micsik, Tamás ; Benedek, Sándor ; Benyó, Z. / A novel method for the detection of apnea and hypopnea events in respiration signals. In: IEEE Transactions on Biomedical Engineering. 2002 ; Vol. 49, No. 9. pp. 936-942.
@article{896a05f0533c49a19e080af0582c38db,
title = "A novel method for the detection of apnea and hypopnea events in respiration signals",
abstract = "The monitoring of breathing dynamics is an essential diagnostic tool in various clinical environments, such as sleep diagnostics, intensive care and neonatal monitoring. This paper introduces an innovative signal classification method that is capable of on-line detection of the presence or absence of normal breathing. Four different artificial neural networks are presented for the recognition of three different patterns in the respiration signals (normal breathing, hypopnea, and apnea). Two networks process the normalized respiration signals directly, while another two use sophisticatedly preprocessed signals. The development of the networks was based on training sets from the polysomnographic records of nine different patients. The detection performance of the networks was tested and compared by using up to 8000 untrained breathing patterns from 16 different patients. The networks which classified the preprocessed respiration signals produced an average detection performance of over 90{\%}. In the light of the moderate computational power used, the presented method is not only viable in clinical polysomnographs and respiration monitors, but also in portable devices.",
keywords = "Classification, Neural networks, Polysomnography, Respiration monitoring, Sleep apnea",
author = "P{\'e}ter V{\'a}rady and Tam{\'a}s Micsik and S{\'a}ndor Benedek and Z. Beny{\'o}",
year = "2002",
month = "9",
doi = "10.1109/TBME.2002.802009",
language = "English",
volume = "49",
pages = "936--942",
journal = "IEEE Transactions on Biomedical Engineering",
issn = "0018-9294",
publisher = "IEEE Computer Society",
number = "9",

}

TY - JOUR

T1 - A novel method for the detection of apnea and hypopnea events in respiration signals

AU - Várady, Péter

AU - Micsik, Tamás

AU - Benedek, Sándor

AU - Benyó, Z.

PY - 2002/9

Y1 - 2002/9

N2 - The monitoring of breathing dynamics is an essential diagnostic tool in various clinical environments, such as sleep diagnostics, intensive care and neonatal monitoring. This paper introduces an innovative signal classification method that is capable of on-line detection of the presence or absence of normal breathing. Four different artificial neural networks are presented for the recognition of three different patterns in the respiration signals (normal breathing, hypopnea, and apnea). Two networks process the normalized respiration signals directly, while another two use sophisticatedly preprocessed signals. The development of the networks was based on training sets from the polysomnographic records of nine different patients. The detection performance of the networks was tested and compared by using up to 8000 untrained breathing patterns from 16 different patients. The networks which classified the preprocessed respiration signals produced an average detection performance of over 90%. In the light of the moderate computational power used, the presented method is not only viable in clinical polysomnographs and respiration monitors, but also in portable devices.

AB - The monitoring of breathing dynamics is an essential diagnostic tool in various clinical environments, such as sleep diagnostics, intensive care and neonatal monitoring. This paper introduces an innovative signal classification method that is capable of on-line detection of the presence or absence of normal breathing. Four different artificial neural networks are presented for the recognition of three different patterns in the respiration signals (normal breathing, hypopnea, and apnea). Two networks process the normalized respiration signals directly, while another two use sophisticatedly preprocessed signals. The development of the networks was based on training sets from the polysomnographic records of nine different patients. The detection performance of the networks was tested and compared by using up to 8000 untrained breathing patterns from 16 different patients. The networks which classified the preprocessed respiration signals produced an average detection performance of over 90%. In the light of the moderate computational power used, the presented method is not only viable in clinical polysomnographs and respiration monitors, but also in portable devices.

KW - Classification

KW - Neural networks

KW - Polysomnography

KW - Respiration monitoring

KW - Sleep apnea

UR - http://www.scopus.com/inward/record.url?scp=0036721383&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036721383&partnerID=8YFLogxK

U2 - 10.1109/TBME.2002.802009

DO - 10.1109/TBME.2002.802009

M3 - Article

VL - 49

SP - 936

EP - 942

JO - IEEE Transactions on Biomedical Engineering

JF - IEEE Transactions on Biomedical Engineering

SN - 0018-9294

IS - 9

ER -