Sleep stage classification with ECG and respiratory effort

Research output: Contribution to journalArticleAcademicpeer-review

49 Citations (Scopus)
4 Downloads (Pure)

Abstract

Automatic sleep stage classification with cardiorespiratory signals has attracted increasing attention. In contrast to the traditional manual scoring based on polysomnography, these signals can be measured using advanced unobtrusive techniques that are currently available, promising the application for personal and continuous home sleep monitoring. This paper describes a methodology for classifying wake, rapid-eye-movement (REM) sleep, and non-REM (NREM) light and deep sleep on a 30 s epoch basis. A total of 142 features were extracted from electrocardiogram and thoracic respiratory effort measured with respiratory inductance plethysmography. To improve the quality of these features, subject-specific Z-score normalization and spline smoothing were used to reduce between-subject and within-subject variability. A modified sequential forward selection feature selector procedure was applied, yielding 80 features while preventing the introduction of bias in the estimation of cross-validation performance. PSG data from 48 healthy adults were used to validate our methods. Using a linear discriminant classifier and a ten-fold cross-validation, we achieved a Cohen's kappa coefficient of 0.49 and an accuracy of 69% in the classification of wake, REM, light, and deep sleep. These values increased to kappa = 0.56 and accuracy = 80% when the classification problem was reduced to three classes, wake, REM sleep, and NREM sleep.
Original languageEnglish
Pages (from-to)2027-2040
Number of pages14
JournalPhysiological Measurement
Volume36
Issue number10
DOIs
Publication statusPublished - 2015

Fingerprint

Sleep Stages
REM Sleep
Electrocardiography
Sleep
Eye movements
Polysomnography
Light
Plethysmography
Eye Movements
Thorax
Inductance
Splines
Feature extraction
Classifiers
Monitoring

Cite this

@article{762656e79e914fd6b88219e8bc5a359c,
title = "Sleep stage classification with ECG and respiratory effort",
abstract = "Automatic sleep stage classification with cardiorespiratory signals has attracted increasing attention. In contrast to the traditional manual scoring based on polysomnography, these signals can be measured using advanced unobtrusive techniques that are currently available, promising the application for personal and continuous home sleep monitoring. This paper describes a methodology for classifying wake, rapid-eye-movement (REM) sleep, and non-REM (NREM) light and deep sleep on a 30 s epoch basis. A total of 142 features were extracted from electrocardiogram and thoracic respiratory effort measured with respiratory inductance plethysmography. To improve the quality of these features, subject-specific Z-score normalization and spline smoothing were used to reduce between-subject and within-subject variability. A modified sequential forward selection feature selector procedure was applied, yielding 80 features while preventing the introduction of bias in the estimation of cross-validation performance. PSG data from 48 healthy adults were used to validate our methods. Using a linear discriminant classifier and a ten-fold cross-validation, we achieved a Cohen's kappa coefficient of 0.49 and an accuracy of 69{\%} in the classification of wake, REM, light, and deep sleep. These values increased to kappa = 0.56 and accuracy = 80{\%} when the classification problem was reduced to three classes, wake, REM sleep, and NREM sleep.",
author = "P. Fonseca and X. Long and M. Radha and R. Haakma and R.M. Aarts and J. Rolink",
year = "2015",
doi = "10.1088/0967-3334/36/10/2027",
language = "English",
volume = "36",
pages = "2027--2040",
journal = "Physiological Measurement",
issn = "0967-3334",
publisher = "Institute of Physics",
number = "10",

}

Sleep stage classification with ECG and respiratory effort. / Fonseca, P.; Long, X.; Radha, M.; Haakma, R.; Aarts, R.M.; Rolink, J.

In: Physiological Measurement, Vol. 36, No. 10, 2015, p. 2027-2040.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Sleep stage classification with ECG and respiratory effort

AU - Fonseca, P.

AU - Long, X.

AU - Radha, M.

AU - Haakma, R.

AU - Aarts, R.M.

AU - Rolink, J.

PY - 2015

Y1 - 2015

N2 - Automatic sleep stage classification with cardiorespiratory signals has attracted increasing attention. In contrast to the traditional manual scoring based on polysomnography, these signals can be measured using advanced unobtrusive techniques that are currently available, promising the application for personal and continuous home sleep monitoring. This paper describes a methodology for classifying wake, rapid-eye-movement (REM) sleep, and non-REM (NREM) light and deep sleep on a 30 s epoch basis. A total of 142 features were extracted from electrocardiogram and thoracic respiratory effort measured with respiratory inductance plethysmography. To improve the quality of these features, subject-specific Z-score normalization and spline smoothing were used to reduce between-subject and within-subject variability. A modified sequential forward selection feature selector procedure was applied, yielding 80 features while preventing the introduction of bias in the estimation of cross-validation performance. PSG data from 48 healthy adults were used to validate our methods. Using a linear discriminant classifier and a ten-fold cross-validation, we achieved a Cohen's kappa coefficient of 0.49 and an accuracy of 69% in the classification of wake, REM, light, and deep sleep. These values increased to kappa = 0.56 and accuracy = 80% when the classification problem was reduced to three classes, wake, REM sleep, and NREM sleep.

AB - Automatic sleep stage classification with cardiorespiratory signals has attracted increasing attention. In contrast to the traditional manual scoring based on polysomnography, these signals can be measured using advanced unobtrusive techniques that are currently available, promising the application for personal and continuous home sleep monitoring. This paper describes a methodology for classifying wake, rapid-eye-movement (REM) sleep, and non-REM (NREM) light and deep sleep on a 30 s epoch basis. A total of 142 features were extracted from electrocardiogram and thoracic respiratory effort measured with respiratory inductance plethysmography. To improve the quality of these features, subject-specific Z-score normalization and spline smoothing were used to reduce between-subject and within-subject variability. A modified sequential forward selection feature selector procedure was applied, yielding 80 features while preventing the introduction of bias in the estimation of cross-validation performance. PSG data from 48 healthy adults were used to validate our methods. Using a linear discriminant classifier and a ten-fold cross-validation, we achieved a Cohen's kappa coefficient of 0.49 and an accuracy of 69% in the classification of wake, REM, light, and deep sleep. These values increased to kappa = 0.56 and accuracy = 80% when the classification problem was reduced to three classes, wake, REM sleep, and NREM sleep.

U2 - 10.1088/0967-3334/36/10/2027

DO - 10.1088/0967-3334/36/10/2027

M3 - Article

C2 - 26289580

VL - 36

SP - 2027

EP - 2040

JO - Physiological Measurement

JF - Physiological Measurement

SN - 0967-3334

IS - 10

ER -