Sleep modelled as a continuous and dynamic process predicts healthy ageing better than traditional sleep scoring.

Age Gaussian mixture model Machine learning Polysomnography Slow wave sleep Spectral analysis

Journal

Sleep medicine
ISSN: 1878-5506
Titre abrégé: Sleep Med
Pays: Netherlands
ID NLM: 100898759

Informations de publication

Date de publication:
01 2021
Historique:
received: 25 08 2020
revised: 24 11 2020
accepted: 27 11 2020
pubmed: 29 12 2020
medline: 22 6 2021
entrez: 28 12 2020
Statut: ppublish

Résumé

In current clinical practice, sleep is manually scored in discrete stages of 30-s duration. We hypothesize that modelling sleep automatically as continuous and dynamic process predicts healthy ageing better than traditional scoring. Sleep electroencephalography of 15 young healthy subjects (aged ≤40 years) was used to train the modelling method. Each 3-s sleep mini-epoch was modelled as a probabilistic combination of wakefulness, light and deep sleep. For 79 healthy sleepers (aged 20-77 years), 15 sleep features were derived from manual traditional scoring (manual features), 7 from the automatic modelling (automatic features) and 24 from a combination of automatic modelling with traditional scoring (combined features). Age was predicted with seven multiple linear regression models with i) manual, ii) automatic, iii) combined, iv) manual + automatic, v) manual + combined, vi) automatic + combined, and vii) manual + automatic + combined sleep features. Using the same seven sleep feature groups, two support vector machine and one random forest classifiers were used to discriminate younger (aged <47 years) from older subjects with fivefold cross-validation. Adjusted coefficients of determination (adj-R The linear model and the classifiers using only manual features achieved the lowest values of adjusted coefficient of determination and classification validation accuracy (adj-R Continuous and dynamic sleep modelling captures healthy ageing better than traditional sleep scoring.

Sections du résumé

BACKGROUND
In current clinical practice, sleep is manually scored in discrete stages of 30-s duration. We hypothesize that modelling sleep automatically as continuous and dynamic process predicts healthy ageing better than traditional scoring.
METHODS
Sleep electroencephalography of 15 young healthy subjects (aged ≤40 years) was used to train the modelling method. Each 3-s sleep mini-epoch was modelled as a probabilistic combination of wakefulness, light and deep sleep. For 79 healthy sleepers (aged 20-77 years), 15 sleep features were derived from manual traditional scoring (manual features), 7 from the automatic modelling (automatic features) and 24 from a combination of automatic modelling with traditional scoring (combined features). Age was predicted with seven multiple linear regression models with i) manual, ii) automatic, iii) combined, iv) manual + automatic, v) manual + combined, vi) automatic + combined, and vii) manual + automatic + combined sleep features. Using the same seven sleep feature groups, two support vector machine and one random forest classifiers were used to discriminate younger (aged <47 years) from older subjects with fivefold cross-validation. Adjusted coefficients of determination (adj-R
RESULTS
The linear model and the classifiers using only manual features achieved the lowest values of adjusted coefficient of determination and classification validation accuracy (adj-R
CONCLUSIONS
Continuous and dynamic sleep modelling captures healthy ageing better than traditional sleep scoring.

Identifiants

pubmed: 33360558
pii: S1389-9457(20)30533-5
doi: 10.1016/j.sleep.2020.11.033
pii:
doi:

Types de publication

Journal Article Research Support, Non-U.S. Gov't

Langues

eng

Sous-ensembles de citation

IM

Pagination

136-146

Subventions

Organisme : Austrian Science Fund FWF
ID : KLI 236
Pays : Austria

Informations de copyright

Copyright © 2020 The Authors. Published by Elsevier B.V. All rights reserved.

Auteurs

Matteo Cesari (M)

Department of Neurology, Medical University of Innsbruck, Innsbruck, Austria. Electronic address: matteo.cesari@i-med.ac.at.

Ambra Stefani (A)

Department of Neurology, Medical University of Innsbruck, Innsbruck, Austria.

Thomas Mitterling (T)

Department of Neurology 1, Kepler University Hospital, Johannes Kepler University, Linz, Austria.

Birgit Frauscher (B)

Analytical Neurophysiology Lab, Montreal Neurological Institute and Hospital, McGill University, Montreal, Canada.

Suzana V Schönwald (SV)

Department of Neurology, Hospital de Clínicas de Porto Alegre, Porto Alegre, Brazil.

Birgit Högl (B)

Department of Neurology, Medical University of Innsbruck, Innsbruck, Austria.

Articles similaires

Humans Male Female Aged Middle Aged
Humans Vitiligo Male Female Adult
Humans Electroencephalography Female Male Middle Aged
Humans Adolescent Mobile Applications United Kingdom Male

Classifications MeSH