On the generalizability of ECG-based obstructive sleep apnea monitoring: merits and limitations of the Apnea-ECG database

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

3 Citations (Scopus)

Abstract

Obstructive sleep apnea syndrome (OSAS) is a sleep disorder that affects a large part of the population and the development of algorithms using cardiovascular features for OSAS monitoring has been an extensively researched topic in the last two decades. Several studies regarding automatic apneic event classification using ECG derived features are based on the public Apnea-ECG database available on PhysioNet. Although this database is an excellent starting point for apnea topic investigations, in our study we show that algorithms for apneic-epochs classification that are successfully trained on this database (sensitivity>85%, false detection rate<20%) perform poorly (sensitivity<55%, false detection rate>40%) in other databases which include patients with a broader spectrum of apneic events and sleep disorders. The reduced performance can be related to the complexity of breathing events, the increased number of non-breathing related sleep events, and the presence of non-OSAS sleep pathologies.
Original languageEnglish
Title of host publication40th International Engineering in Medicine and Biology Conference
Number of pages4
Publication statusPublished - 22 Jul 2018
Event40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society= (EMBC 2018) - Hawaii Convention Center, Honolulu, United States
Duration: 18 Jul 201821 Jul 2018
Conference number: 40
https://embc.embs.org/2018/

Conference

Conference40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society= (EMBC 2018)
Abbreviated titleEMBC 2018
CountryUnited States
CityHonolulu
Period18/07/1821/07/18
Other"Learning from the Past, Looking to the Future"
Internet address

Fingerprint

Polysomnography
Obstructive Sleep Apnea
Apnea
Electrocardiography
Databases
Sleep
Sleep Apnea Syndromes
Respiration
Pathology
Population
Sleep Wake Disorders

Cite this

@inproceedings{b662c7a07d7d4d29ba89a65120d4daac,
title = "On the generalizability of ECG-based obstructive sleep apnea monitoring: merits and limitations of the Apnea-ECG database",
abstract = "Obstructive sleep apnea syndrome (OSAS) is a sleep disorder that affects a large part of the population and the development of algorithms using cardiovascular features for OSAS monitoring has been an extensively researched topic in the last two decades. Several studies regarding automatic apneic event classification using ECG derived features are based on the public Apnea-ECG database available on PhysioNet. Although this database is an excellent starting point for apnea topic investigations, in our study we show that algorithms for apneic-epochs classification that are successfully trained on this database (sensitivity>85{\%}, false detection rate<20{\%}) perform poorly (sensitivity<55{\%}, false detection rate>40{\%}) in other databases which include patients with a broader spectrum of apneic events and sleep disorders. The reduced performance can be related to the complexity of breathing events, the increased number of non-breathing related sleep events, and the presence of non-OSAS sleep pathologies.",
author = "G. Papini and P. Fonseca and Jenny Margarito and {van Gilst}, M.M. and S. Overeem and J.W.M. Bergmans and R. Vullings",
year = "2018",
month = "7",
day = "22",
language = "English",
booktitle = "40th International Engineering in Medicine and Biology Conference",

}

Papini, G, Fonseca, P, Margarito, J, van Gilst, MM, Overeem, S, Bergmans, JWM & Vullings, R 2018, On the generalizability of ECG-based obstructive sleep apnea monitoring: merits and limitations of the Apnea-ECG database. in 40th International Engineering in Medicine and Biology Conference. 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society= (EMBC 2018), Honolulu, United States, 18/07/18.

On the generalizability of ECG-based obstructive sleep apnea monitoring: merits and limitations of the Apnea-ECG database. / Papini, G.; Fonseca, P.; Margarito, Jenny; van Gilst, M.M.; Overeem, S.; Bergmans, J.W.M.; Vullings, R.

40th International Engineering in Medicine and Biology Conference. 2018.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - On the generalizability of ECG-based obstructive sleep apnea monitoring: merits and limitations of the Apnea-ECG database

AU - Papini, G.

AU - Fonseca, P.

AU - Margarito, Jenny

AU - van Gilst, M.M.

AU - Overeem, S.

AU - Bergmans, J.W.M.

AU - Vullings, R.

PY - 2018/7/22

Y1 - 2018/7/22

N2 - Obstructive sleep apnea syndrome (OSAS) is a sleep disorder that affects a large part of the population and the development of algorithms using cardiovascular features for OSAS monitoring has been an extensively researched topic in the last two decades. Several studies regarding automatic apneic event classification using ECG derived features are based on the public Apnea-ECG database available on PhysioNet. Although this database is an excellent starting point for apnea topic investigations, in our study we show that algorithms for apneic-epochs classification that are successfully trained on this database (sensitivity>85%, false detection rate<20%) perform poorly (sensitivity<55%, false detection rate>40%) in other databases which include patients with a broader spectrum of apneic events and sleep disorders. The reduced performance can be related to the complexity of breathing events, the increased number of non-breathing related sleep events, and the presence of non-OSAS sleep pathologies.

AB - Obstructive sleep apnea syndrome (OSAS) is a sleep disorder that affects a large part of the population and the development of algorithms using cardiovascular features for OSAS monitoring has been an extensively researched topic in the last two decades. Several studies regarding automatic apneic event classification using ECG derived features are based on the public Apnea-ECG database available on PhysioNet. Although this database is an excellent starting point for apnea topic investigations, in our study we show that algorithms for apneic-epochs classification that are successfully trained on this database (sensitivity>85%, false detection rate<20%) perform poorly (sensitivity<55%, false detection rate>40%) in other databases which include patients with a broader spectrum of apneic events and sleep disorders. The reduced performance can be related to the complexity of breathing events, the increased number of non-breathing related sleep events, and the presence of non-OSAS sleep pathologies.

M3 - Conference contribution

BT - 40th International Engineering in Medicine and Biology Conference

ER -