ECG-Derived Respiration Using a Real-Time QRS Detector Based on Empirical Mode Decomposition

Christina Kozia, Randa Herzallah, David Lowe

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

Abstract

Respiration Rate (RR) is an important physiological
indicator and plays a major role in health deterioration monitoring.
Despite that, it has been neglected in hospital wards due to
inadequate nursing skills and insufficient equipment. ECG signal,
which is always monitored in a clinical setting, is modulated
by respiration which renders it a highly enticing mean for the
automatic RR estimation. In addition, accurate QRS detection is
pivotal to RR estimation from the ECG signal. The investigation
of QRS complexes is a continuing concern in ECG analysis
because current methods are still inaccurate and miss heart
beats. This paper presents a frequency domain RR estimation
method which uses a novel real-time QRS detector based on
Empirical Mode Decomposition (EMD). Another novelty of the
proposed work stems from the RR estimation in the frequency
domain as opposed to some of the current methods which rely
on a time domain analysis. As will be shown later, the RR
extraction in the frequency domain provides more accurate
results compared to the time domain methods. Moreover, our
novel QRS detector uses an adaptive threshold over a sliding
window and differentiates large Q- from R-peaks, facilitating a
more accurate RR estimation. The performance of our methods
was tested on real data from Capnobase dataset. An average
mean absolute error of less than 0.5 breath per minute was
achieved using our frequency domain method, compared to
6 breaths per minute of the time domain analysis. Moreover,
our modified QRS detector shows comparable results to other
published methods, achieving a detection rate over 99.80%.
Original languageEnglish
Title of host publication12th International Conference on Signal Processing and Communication Systems. ICSPCS 2018
EditorsTadeusz A Wysocki, Beata J Wysocki
PublisherIEEE
ISBN (Electronic)978-1-5386-5602-0
ISBN (Print)978-1-5386-5603-7
DOIs
Publication statusPublished - 4 Feb 2019
Event12th International Conference on Signal Processing and Communication Systems, ICSPCS 2018 - Cairns, Australia
Duration: 17 Dec 201819 Dec 2018

Conference

Conference12th International Conference on Signal Processing and Communication Systems, ICSPCS 2018
CountryAustralia
CityCairns
Period17/12/1819/12/18

Fingerprint

Electrocardiography
Detectors
Decomposition
Time domain analysis
Nursing
Deterioration
Health
Monitoring

Bibliographical note

© 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Keywords

  • ECG-derived-respiration
  • Empirical Mode Decomposition (EMD)
  • Frequency domain analysis
  • Local Signal Energy
  • R-peak detection

Cite this

Kozia, C., Herzallah, R., & Lowe, D. (2019). ECG-Derived Respiration Using a Real-Time QRS Detector Based on Empirical Mode Decomposition. In T. A. Wysocki, & B. J. Wysocki (Eds.), 12th International Conference on Signal Processing and Communication Systems. ICSPCS 2018 [8631760] IEEE. https://doi.org/10.1109/ICSPCS.2018.8631760
Kozia, Christina ; Herzallah, Randa ; Lowe, David. / ECG-Derived Respiration Using a Real-Time QRS Detector Based on Empirical Mode Decomposition. 12th International Conference on Signal Processing and Communication Systems. ICSPCS 2018. editor / Tadeusz A Wysocki ; Beata J Wysocki. IEEE, 2019.
@inproceedings{1eca1dd70fbf4858b6b8f592d1678e84,
title = "ECG-Derived Respiration Using a Real-Time QRS Detector Based on Empirical Mode Decomposition",
abstract = "Respiration Rate (RR) is an important physiologicalindicator and plays a major role in health deterioration monitoring.Despite that, it has been neglected in hospital wards due toinadequate nursing skills and insufficient equipment. ECG signal,which is always monitored in a clinical setting, is modulatedby respiration which renders it a highly enticing mean for theautomatic RR estimation. In addition, accurate QRS detection ispivotal to RR estimation from the ECG signal. The investigationof QRS complexes is a continuing concern in ECG analysisbecause current methods are still inaccurate and miss heartbeats. This paper presents a frequency domain RR estimationmethod which uses a novel real-time QRS detector based onEmpirical Mode Decomposition (EMD). Another novelty of theproposed work stems from the RR estimation in the frequencydomain as opposed to some of the current methods which relyon a time domain analysis. As will be shown later, the RRextraction in the frequency domain provides more accurateresults compared to the time domain methods. Moreover, ournovel QRS detector uses an adaptive threshold over a slidingwindow and differentiates large Q- from R-peaks, facilitating amore accurate RR estimation. The performance of our methodswas tested on real data from Capnobase dataset. An averagemean absolute error of less than 0.5 breath per minute wasachieved using our frequency domain method, compared to6 breaths per minute of the time domain analysis. Moreover,our modified QRS detector shows comparable results to otherpublished methods, achieving a detection rate over 99.80{\%}.",
keywords = "ECG-derived-respiration, Empirical Mode Decomposition (EMD), Frequency domain analysis, Local Signal Energy, R-peak detection",
author = "Christina Kozia and Randa Herzallah and David Lowe",
note = "{\circledC} 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.",
year = "2019",
month = "2",
day = "4",
doi = "10.1109/ICSPCS.2018.8631760",
language = "English",
isbn = "978-1-5386-5603-7",
editor = "Wysocki, {Tadeusz A} and Wysocki, {Beata J}",
booktitle = "12th International Conference on Signal Processing and Communication Systems. ICSPCS 2018",
publisher = "IEEE",
address = "United States",

}

Kozia, C, Herzallah, R & Lowe, D 2019, ECG-Derived Respiration Using a Real-Time QRS Detector Based on Empirical Mode Decomposition. in TA Wysocki & BJ Wysocki (eds), 12th International Conference on Signal Processing and Communication Systems. ICSPCS 2018., 8631760, IEEE, 12th International Conference on Signal Processing and Communication Systems, ICSPCS 2018, Cairns, Australia, 17/12/18. https://doi.org/10.1109/ICSPCS.2018.8631760

ECG-Derived Respiration Using a Real-Time QRS Detector Based on Empirical Mode Decomposition. / Kozia, Christina; Herzallah, Randa; Lowe, David.

12th International Conference on Signal Processing and Communication Systems. ICSPCS 2018. ed. / Tadeusz A Wysocki; Beata J Wysocki. IEEE, 2019. 8631760.

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

TY - GEN

T1 - ECG-Derived Respiration Using a Real-Time QRS Detector Based on Empirical Mode Decomposition

AU - Kozia, Christina

AU - Herzallah, Randa

AU - Lowe, David

N1 - © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

PY - 2019/2/4

Y1 - 2019/2/4

N2 - Respiration Rate (RR) is an important physiologicalindicator and plays a major role in health deterioration monitoring.Despite that, it has been neglected in hospital wards due toinadequate nursing skills and insufficient equipment. ECG signal,which is always monitored in a clinical setting, is modulatedby respiration which renders it a highly enticing mean for theautomatic RR estimation. In addition, accurate QRS detection ispivotal to RR estimation from the ECG signal. The investigationof QRS complexes is a continuing concern in ECG analysisbecause current methods are still inaccurate and miss heartbeats. This paper presents a frequency domain RR estimationmethod which uses a novel real-time QRS detector based onEmpirical Mode Decomposition (EMD). Another novelty of theproposed work stems from the RR estimation in the frequencydomain as opposed to some of the current methods which relyon a time domain analysis. As will be shown later, the RRextraction in the frequency domain provides more accurateresults compared to the time domain methods. Moreover, ournovel QRS detector uses an adaptive threshold over a slidingwindow and differentiates large Q- from R-peaks, facilitating amore accurate RR estimation. The performance of our methodswas tested on real data from Capnobase dataset. An averagemean absolute error of less than 0.5 breath per minute wasachieved using our frequency domain method, compared to6 breaths per minute of the time domain analysis. Moreover,our modified QRS detector shows comparable results to otherpublished methods, achieving a detection rate over 99.80%.

AB - Respiration Rate (RR) is an important physiologicalindicator and plays a major role in health deterioration monitoring.Despite that, it has been neglected in hospital wards due toinadequate nursing skills and insufficient equipment. ECG signal,which is always monitored in a clinical setting, is modulatedby respiration which renders it a highly enticing mean for theautomatic RR estimation. In addition, accurate QRS detection ispivotal to RR estimation from the ECG signal. The investigationof QRS complexes is a continuing concern in ECG analysisbecause current methods are still inaccurate and miss heartbeats. This paper presents a frequency domain RR estimationmethod which uses a novel real-time QRS detector based onEmpirical Mode Decomposition (EMD). Another novelty of theproposed work stems from the RR estimation in the frequencydomain as opposed to some of the current methods which relyon a time domain analysis. As will be shown later, the RRextraction in the frequency domain provides more accurateresults compared to the time domain methods. Moreover, ournovel QRS detector uses an adaptive threshold over a slidingwindow and differentiates large Q- from R-peaks, facilitating amore accurate RR estimation. The performance of our methodswas tested on real data from Capnobase dataset. An averagemean absolute error of less than 0.5 breath per minute wasachieved using our frequency domain method, compared to6 breaths per minute of the time domain analysis. Moreover,our modified QRS detector shows comparable results to otherpublished methods, achieving a detection rate over 99.80%.

KW - ECG-derived-respiration

KW - Empirical Mode Decomposition (EMD)

KW - Frequency domain analysis

KW - Local Signal Energy

KW - R-peak detection

UR - https://www.dspcs-witsp.com/icspcs_2018/index.html

UR - https://ieeexplore.ieee.org/document/8631760?tp=&arnumber=8631760&contentType=Conferences&dld=YXN0b24uYWMudWs%3D&source=SEARCHALERT

UR - http://www.scopus.com/inward/record.url?scp=85062877799&partnerID=8YFLogxK

U2 - 10.1109/ICSPCS.2018.8631760

DO - 10.1109/ICSPCS.2018.8631760

M3 - Conference contribution

SN - 978-1-5386-5603-7

BT - 12th International Conference on Signal Processing and Communication Systems. ICSPCS 2018

A2 - Wysocki, Tadeusz A

A2 - Wysocki, Beata J

PB - IEEE

ER -

Kozia C, Herzallah R, Lowe D. ECG-Derived Respiration Using a Real-Time QRS Detector Based on Empirical Mode Decomposition. In Wysocki TA, Wysocki BJ, editors, 12th International Conference on Signal Processing and Communication Systems. ICSPCS 2018. IEEE. 2019. 8631760 https://doi.org/10.1109/ICSPCS.2018.8631760