Automatic Sleep Arousal Detection Using State Distance Analysis in Phase Space

Morteza Zabihi*, Ali Bahrami Rad, Simo Sarkka, Serkan Kiranyaz, Aggelos K Katsaggelos, Moncef Gabbouj

*Corresponding author for this work

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

1 Citation (Scopus)

Abstract

Defective sleep arousal can contribute to significant sleep-related injuries and affect the quality of life. Investigating the arousal process is a challenging task as most of such events may be associated with subtle electrophysiological indications. Thus, developing an accurate model is an essential step toward the diagnosis and assessment of arousals. Here we introduce a novel approach for automatic arousal detection inspired by the states' recurrences in nonlinear dynamics. We first show how the states distance matrices of a complex system can be reconstructed to decrease the effect of false neighbors. Then, we use a convolutional neural network for probing the correlated structures inside the distance matrices with the arousal occurrences. Contrary to earlier studies in the literature, the proposed approach focuses on the dynamic behavior of polysomnography recordings rather than frequency analysis. The proposed approach is evaluated on the training dataset in a 3-fold cross-validation scheme and achieved an average of 19.20% and 78.57% for the area under the precision-recall (AUPRC) and area under the ROC curves, respectively. The overall AUPRC on the unseen test dataset is 19%.

Original languageEnglish (US)
Title of host publicationComputing in Cardiology Conference, CinC 2018
PublisherIEEE Computer Society
ISBN (Electronic)9781728109589
DOIs
StatePublished - Sep 1 2018
Event45th Computing in Cardiology Conference, CinC 2018 - Maastricht, Netherlands
Duration: Sep 23 2018Sep 26 2018

Publication series

NameComputing in Cardiology
Volume2018-September
ISSN (Print)2325-8861
ISSN (Electronic)2325-887X

Conference

Conference45th Computing in Cardiology Conference, CinC 2018
CountryNetherlands
CityMaastricht
Period9/23/189/26/18

Fingerprint

Arousal
Sleep
Large scale systems
Neural networks
Nonlinear Dynamics
Polysomnography
ROC Curve
Area Under Curve
Quality of Life
Recurrence
Wounds and Injuries
Datasets

ASJC Scopus subject areas

  • Computer Science(all)
  • Cardiology and Cardiovascular Medicine

Cite this

Zabihi, M., Rad, A. B., Sarkka, S., Kiranyaz, S., Katsaggelos, A. K., & Gabbouj, M. (2018). Automatic Sleep Arousal Detection Using State Distance Analysis in Phase Space. In Computing in Cardiology Conference, CinC 2018 [8744012] (Computing in Cardiology; Vol. 2018-September). IEEE Computer Society. https://doi.org/10.22489/CinC.2018.257
Zabihi, Morteza ; Rad, Ali Bahrami ; Sarkka, Simo ; Kiranyaz, Serkan ; Katsaggelos, Aggelos K ; Gabbouj, Moncef. / Automatic Sleep Arousal Detection Using State Distance Analysis in Phase Space. Computing in Cardiology Conference, CinC 2018. IEEE Computer Society, 2018. (Computing in Cardiology).
@inproceedings{243f10c3b0d54c1ba96db4a704f3629e,
title = "Automatic Sleep Arousal Detection Using State Distance Analysis in Phase Space",
abstract = "Defective sleep arousal can contribute to significant sleep-related injuries and affect the quality of life. Investigating the arousal process is a challenging task as most of such events may be associated with subtle electrophysiological indications. Thus, developing an accurate model is an essential step toward the diagnosis and assessment of arousals. Here we introduce a novel approach for automatic arousal detection inspired by the states' recurrences in nonlinear dynamics. We first show how the states distance matrices of a complex system can be reconstructed to decrease the effect of false neighbors. Then, we use a convolutional neural network for probing the correlated structures inside the distance matrices with the arousal occurrences. Contrary to earlier studies in the literature, the proposed approach focuses on the dynamic behavior of polysomnography recordings rather than frequency analysis. The proposed approach is evaluated on the training dataset in a 3-fold cross-validation scheme and achieved an average of 19.20{\%} and 78.57{\%} for the area under the precision-recall (AUPRC) and area under the ROC curves, respectively. The overall AUPRC on the unseen test dataset is 19{\%}.",
author = "Morteza Zabihi and Rad, {Ali Bahrami} and Simo Sarkka and Serkan Kiranyaz and Katsaggelos, {Aggelos K} and Moncef Gabbouj",
year = "2018",
month = "9",
day = "1",
doi = "10.22489/CinC.2018.257",
language = "English (US)",
series = "Computing in Cardiology",
publisher = "IEEE Computer Society",
booktitle = "Computing in Cardiology Conference, CinC 2018",
address = "United States",

}

Zabihi, M, Rad, AB, Sarkka, S, Kiranyaz, S, Katsaggelos, AK & Gabbouj, M 2018, Automatic Sleep Arousal Detection Using State Distance Analysis in Phase Space. in Computing in Cardiology Conference, CinC 2018., 8744012, Computing in Cardiology, vol. 2018-September, IEEE Computer Society, 45th Computing in Cardiology Conference, CinC 2018, Maastricht, Netherlands, 9/23/18. https://doi.org/10.22489/CinC.2018.257

Automatic Sleep Arousal Detection Using State Distance Analysis in Phase Space. / Zabihi, Morteza; Rad, Ali Bahrami; Sarkka, Simo; Kiranyaz, Serkan; Katsaggelos, Aggelos K; Gabbouj, Moncef.

Computing in Cardiology Conference, CinC 2018. IEEE Computer Society, 2018. 8744012 (Computing in Cardiology; Vol. 2018-September).

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

TY - GEN

T1 - Automatic Sleep Arousal Detection Using State Distance Analysis in Phase Space

AU - Zabihi, Morteza

AU - Rad, Ali Bahrami

AU - Sarkka, Simo

AU - Kiranyaz, Serkan

AU - Katsaggelos, Aggelos K

AU - Gabbouj, Moncef

PY - 2018/9/1

Y1 - 2018/9/1

N2 - Defective sleep arousal can contribute to significant sleep-related injuries and affect the quality of life. Investigating the arousal process is a challenging task as most of such events may be associated with subtle electrophysiological indications. Thus, developing an accurate model is an essential step toward the diagnosis and assessment of arousals. Here we introduce a novel approach for automatic arousal detection inspired by the states' recurrences in nonlinear dynamics. We first show how the states distance matrices of a complex system can be reconstructed to decrease the effect of false neighbors. Then, we use a convolutional neural network for probing the correlated structures inside the distance matrices with the arousal occurrences. Contrary to earlier studies in the literature, the proposed approach focuses on the dynamic behavior of polysomnography recordings rather than frequency analysis. The proposed approach is evaluated on the training dataset in a 3-fold cross-validation scheme and achieved an average of 19.20% and 78.57% for the area under the precision-recall (AUPRC) and area under the ROC curves, respectively. The overall AUPRC on the unseen test dataset is 19%.

AB - Defective sleep arousal can contribute to significant sleep-related injuries and affect the quality of life. Investigating the arousal process is a challenging task as most of such events may be associated with subtle electrophysiological indications. Thus, developing an accurate model is an essential step toward the diagnosis and assessment of arousals. Here we introduce a novel approach for automatic arousal detection inspired by the states' recurrences in nonlinear dynamics. We first show how the states distance matrices of a complex system can be reconstructed to decrease the effect of false neighbors. Then, we use a convolutional neural network for probing the correlated structures inside the distance matrices with the arousal occurrences. Contrary to earlier studies in the literature, the proposed approach focuses on the dynamic behavior of polysomnography recordings rather than frequency analysis. The proposed approach is evaluated on the training dataset in a 3-fold cross-validation scheme and achieved an average of 19.20% and 78.57% for the area under the precision-recall (AUPRC) and area under the ROC curves, respectively. The overall AUPRC on the unseen test dataset is 19%.

UR - http://www.scopus.com/inward/record.url?scp=85068778890&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068778890&partnerID=8YFLogxK

U2 - 10.22489/CinC.2018.257

DO - 10.22489/CinC.2018.257

M3 - Conference contribution

T3 - Computing in Cardiology

BT - Computing in Cardiology Conference, CinC 2018

PB - IEEE Computer Society

ER -

Zabihi M, Rad AB, Sarkka S, Kiranyaz S, Katsaggelos AK, Gabbouj M. Automatic Sleep Arousal Detection Using State Distance Analysis in Phase Space. In Computing in Cardiology Conference, CinC 2018. IEEE Computer Society. 2018. 8744012. (Computing in Cardiology). https://doi.org/10.22489/CinC.2018.257