Sleep and wake classification with actigraphy and respiratory effort using dynamic warping

Research output: Contribution to journalArticleAcademicpeer-review

47 Citations (Scopus)
1 Downloads (Pure)

Abstract

This paper proposes the use of dynamic warping (DW) methods for improving automatic sleep and wake classification using actigraphy and respiratory effort. DW is an algorithm that finds an optimal non-linear alignment between two series allowing scaling and shifting. It is widely used to quantify (dis)similarity between two series. To compare the respiratory effort between sleep and wake states by means of (dis)similarity, we constructed two novel features based on DW. For a given epoch of a respiratory effort recording, the features search for the optimally aligned epoch within the same recording in time and frequency domain. This is expected to yield a high (or low) similarity score when this epoch is sleep (or wake). Since the comparison occurs throughout the entire-night recording of a subject, it may reduce the effects of within- and between-subject variations of respiratory effort, and thus help discriminate between sleep and wake states. The DW-based features were evaluated using a Linear Discriminant classifier on a data set of 15 healthy subjects. Results show that the DW-based features can provide a Cohen’s Kappa coefficient of agreement Kappa = 0.59 which is significantly higher than the existing respiratory-based features and is comparable to actigraphy. After combining the actigraphy and the DW-based features, the classifier achieved a of 0.66 and an overall accuracy of 95.7%, outperforming an earlier actigraphy- and respiratory-based feature set ( Kappa = 0.62). The results are also comparable with those obtained using an actigraphy- and cardiorespiratory-based feature set but have the important advantage that they do not require an ECG signal to be recorded.
Original languageEnglish
Pages (from-to)1272-1284
Number of pages13
JournalIEEE Journal of Biomedical and Health Informatics
Volume18
Issue number4
DOIs
Publication statusPublished - 2014

Fingerprint

Actigraphy
Sleep
Classifiers
Electrocardiography
Healthy Volunteers

Cite this

@article{30aa5bc0bc034546aa5d6b8b35980d9b,
title = "Sleep and wake classification with actigraphy and respiratory effort using dynamic warping",
abstract = "This paper proposes the use of dynamic warping (DW) methods for improving automatic sleep and wake classification using actigraphy and respiratory effort. DW is an algorithm that finds an optimal non-linear alignment between two series allowing scaling and shifting. It is widely used to quantify (dis)similarity between two series. To compare the respiratory effort between sleep and wake states by means of (dis)similarity, we constructed two novel features based on DW. For a given epoch of a respiratory effort recording, the features search for the optimally aligned epoch within the same recording in time and frequency domain. This is expected to yield a high (or low) similarity score when this epoch is sleep (or wake). Since the comparison occurs throughout the entire-night recording of a subject, it may reduce the effects of within- and between-subject variations of respiratory effort, and thus help discriminate between sleep and wake states. The DW-based features were evaluated using a Linear Discriminant classifier on a data set of 15 healthy subjects. Results show that the DW-based features can provide a Cohen’s Kappa coefficient of agreement Kappa = 0.59 which is significantly higher than the existing respiratory-based features and is comparable to actigraphy. After combining the actigraphy and the DW-based features, the classifier achieved a of 0.66 and an overall accuracy of 95.7{\%}, outperforming an earlier actigraphy- and respiratory-based feature set ( Kappa = 0.62). The results are also comparable with those obtained using an actigraphy- and cardiorespiratory-based feature set but have the important advantage that they do not require an ECG signal to be recorded.",
author = "X. Long and P. Fonseca and J. Foussier and R. Haakma and R.M. Aarts",
year = "2014",
doi = "10.1109/JBHI.2013.2284610",
language = "English",
volume = "18",
pages = "1272--1284",
journal = "IEEE Journal of Biomedical and Health Informatics",
issn = "2168-2194",
publisher = "Institute of Electrical and Electronics Engineers",
number = "4",

}

Sleep and wake classification with actigraphy and respiratory effort using dynamic warping. / Long, X.; Fonseca, P.; Foussier, J.; Haakma, R.; Aarts, R.M.

In: IEEE Journal of Biomedical and Health Informatics, Vol. 18, No. 4, 2014, p. 1272-1284.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Sleep and wake classification with actigraphy and respiratory effort using dynamic warping

AU - Long, X.

AU - Fonseca, P.

AU - Foussier, J.

AU - Haakma, R.

AU - Aarts, R.M.

PY - 2014

Y1 - 2014

N2 - This paper proposes the use of dynamic warping (DW) methods for improving automatic sleep and wake classification using actigraphy and respiratory effort. DW is an algorithm that finds an optimal non-linear alignment between two series allowing scaling and shifting. It is widely used to quantify (dis)similarity between two series. To compare the respiratory effort between sleep and wake states by means of (dis)similarity, we constructed two novel features based on DW. For a given epoch of a respiratory effort recording, the features search for the optimally aligned epoch within the same recording in time and frequency domain. This is expected to yield a high (or low) similarity score when this epoch is sleep (or wake). Since the comparison occurs throughout the entire-night recording of a subject, it may reduce the effects of within- and between-subject variations of respiratory effort, and thus help discriminate between sleep and wake states. The DW-based features were evaluated using a Linear Discriminant classifier on a data set of 15 healthy subjects. Results show that the DW-based features can provide a Cohen’s Kappa coefficient of agreement Kappa = 0.59 which is significantly higher than the existing respiratory-based features and is comparable to actigraphy. After combining the actigraphy and the DW-based features, the classifier achieved a of 0.66 and an overall accuracy of 95.7%, outperforming an earlier actigraphy- and respiratory-based feature set ( Kappa = 0.62). The results are also comparable with those obtained using an actigraphy- and cardiorespiratory-based feature set but have the important advantage that they do not require an ECG signal to be recorded.

AB - This paper proposes the use of dynamic warping (DW) methods for improving automatic sleep and wake classification using actigraphy and respiratory effort. DW is an algorithm that finds an optimal non-linear alignment between two series allowing scaling and shifting. It is widely used to quantify (dis)similarity between two series. To compare the respiratory effort between sleep and wake states by means of (dis)similarity, we constructed two novel features based on DW. For a given epoch of a respiratory effort recording, the features search for the optimally aligned epoch within the same recording in time and frequency domain. This is expected to yield a high (or low) similarity score when this epoch is sleep (or wake). Since the comparison occurs throughout the entire-night recording of a subject, it may reduce the effects of within- and between-subject variations of respiratory effort, and thus help discriminate between sleep and wake states. The DW-based features were evaluated using a Linear Discriminant classifier on a data set of 15 healthy subjects. Results show that the DW-based features can provide a Cohen’s Kappa coefficient of agreement Kappa = 0.59 which is significantly higher than the existing respiratory-based features and is comparable to actigraphy. After combining the actigraphy and the DW-based features, the classifier achieved a of 0.66 and an overall accuracy of 95.7%, outperforming an earlier actigraphy- and respiratory-based feature set ( Kappa = 0.62). The results are also comparable with those obtained using an actigraphy- and cardiorespiratory-based feature set but have the important advantage that they do not require an ECG signal to be recorded.

U2 - 10.1109/JBHI.2013.2284610

DO - 10.1109/JBHI.2013.2284610

M3 - Article

C2 - 24108754

VL - 18

SP - 1272

EP - 1284

JO - IEEE Journal of Biomedical and Health Informatics

JF - IEEE Journal of Biomedical and Health Informatics

SN - 2168-2194

IS - 4

ER -