Sleep staging from heart rate variability: Time-varying spectral features and hidden markov models

Martin Oswaldo Mendez, Matteo Matteucci, Vincenza Castronovo, Luigi Ferini-Strambi, Sergio Cerutti, Anna Maria Bianchi

Research output: Contribution to journalArticle

Abstract

An alternative DSS which models the behaviour of the Heart Rate Variability (HRV) signal linked to stable (NREM) and instable (REM) cerebral waves during sleep and a probabilistic model of the sleep stages transitions for decision was developed. Time-Varying Autoregressive Models (TVAMs) were used as feature extractor while Hidden Markov Models (HMM) was used as time series classifier. 24 full polysomnography recordings from healthy sleepers were used for the analysis and those were separated in two sets of 12 each: training and test set. The classification performance for the test set was specificity = 0.851, accuracy = 0.793 and sensitivity = 0.702.

Original languageEnglish
Pages (from-to)246-263
Number of pages18
JournalInternational Journal of Biomedical Engineering and Technology
Volume3
Issue number3-4
DOIs
Publication statusPublished - Apr 2010

Fingerprint

Hidden Markov models
Time series
Classifiers
Sleep
Statistical Models

Keywords

  • Decision-support systems
  • DSS
  • Heart rate variability
  • Hidden Markov model
  • HMM
  • HRV
  • Sleep
  • Sleep staging
  • Time-varying analysis

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

Sleep staging from heart rate variability : Time-varying spectral features and hidden markov models. / Mendez, Martin Oswaldo; Matteucci, Matteo; Castronovo, Vincenza; Ferini-Strambi, Luigi; Cerutti, Sergio; Bianchi, Anna Maria.

In: International Journal of Biomedical Engineering and Technology, Vol. 3, No. 3-4, 04.2010, p. 246-263.

Research output: Contribution to journalArticle

Mendez, Martin Oswaldo ; Matteucci, Matteo ; Castronovo, Vincenza ; Ferini-Strambi, Luigi ; Cerutti, Sergio ; Bianchi, Anna Maria. / Sleep staging from heart rate variability : Time-varying spectral features and hidden markov models. In: International Journal of Biomedical Engineering and Technology. 2010 ; Vol. 3, No. 3-4. pp. 246-263.
@article{2ef2cea3d03f48c6b9bf0e43c845e766,
title = "Sleep staging from heart rate variability: Time-varying spectral features and hidden markov models",
abstract = "An alternative DSS which models the behaviour of the Heart Rate Variability (HRV) signal linked to stable (NREM) and instable (REM) cerebral waves during sleep and a probabilistic model of the sleep stages transitions for decision was developed. Time-Varying Autoregressive Models (TVAMs) were used as feature extractor while Hidden Markov Models (HMM) was used as time series classifier. 24 full polysomnography recordings from healthy sleepers were used for the analysis and those were separated in two sets of 12 each: training and test set. The classification performance for the test set was specificity = 0.851, accuracy = 0.793 and sensitivity = 0.702.",
keywords = "Decision-support systems, DSS, Heart rate variability, Hidden Markov model, HMM, HRV, Sleep, Sleep staging, Time-varying analysis",
author = "Mendez, {Martin Oswaldo} and Matteo Matteucci and Vincenza Castronovo and Luigi Ferini-Strambi and Sergio Cerutti and Bianchi, {Anna Maria}",
year = "2010",
month = "4",
doi = "10.1504/IJBET.2010.032695",
language = "English",
volume = "3",
pages = "246--263",
journal = "International Journal of Biomedical Engineering and Technology",
issn = "1752-6418",
publisher = "Inderscience Enterprises Ltd",
number = "3-4",

}

TY - JOUR

T1 - Sleep staging from heart rate variability

T2 - Time-varying spectral features and hidden markov models

AU - Mendez, Martin Oswaldo

AU - Matteucci, Matteo

AU - Castronovo, Vincenza

AU - Ferini-Strambi, Luigi

AU - Cerutti, Sergio

AU - Bianchi, Anna Maria

PY - 2010/4

Y1 - 2010/4

N2 - An alternative DSS which models the behaviour of the Heart Rate Variability (HRV) signal linked to stable (NREM) and instable (REM) cerebral waves during sleep and a probabilistic model of the sleep stages transitions for decision was developed. Time-Varying Autoregressive Models (TVAMs) were used as feature extractor while Hidden Markov Models (HMM) was used as time series classifier. 24 full polysomnography recordings from healthy sleepers were used for the analysis and those were separated in two sets of 12 each: training and test set. The classification performance for the test set was specificity = 0.851, accuracy = 0.793 and sensitivity = 0.702.

AB - An alternative DSS which models the behaviour of the Heart Rate Variability (HRV) signal linked to stable (NREM) and instable (REM) cerebral waves during sleep and a probabilistic model of the sleep stages transitions for decision was developed. Time-Varying Autoregressive Models (TVAMs) were used as feature extractor while Hidden Markov Models (HMM) was used as time series classifier. 24 full polysomnography recordings from healthy sleepers were used for the analysis and those were separated in two sets of 12 each: training and test set. The classification performance for the test set was specificity = 0.851, accuracy = 0.793 and sensitivity = 0.702.

KW - Decision-support systems

KW - DSS

KW - Heart rate variability

KW - Hidden Markov model

KW - HMM

KW - HRV

KW - Sleep

KW - Sleep staging

KW - Time-varying analysis

UR - http://www.scopus.com/inward/record.url?scp=77953163689&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77953163689&partnerID=8YFLogxK

U2 - 10.1504/IJBET.2010.032695

DO - 10.1504/IJBET.2010.032695

M3 - Article

AN - SCOPUS:77953163689

VL - 3

SP - 246

EP - 263

JO - International Journal of Biomedical Engineering and Technology

JF - International Journal of Biomedical Engineering and Technology

SN - 1752-6418

IS - 3-4

ER -