Sleep stage classification from heart-rate variability using long short-term memory neural networks

Mustafa Radha (Corresponding author), Pedro Fonseca, Arnaud Moreau, Marco Ross, Andreas Cerny, Peter Anderer, Xi Long, Ronald Aarts

Research output: Contribution to journalArticleAcademicpeer-review

71 Downloads (Pure)

Abstract

Automated sleep stage classification using heart rate variability (HRV) may provide an ergonomic and low-cost alternative to gold standard polysomnography, creating possibilities for unobtrusive home-based sleep monitoring. Current methods however are limited in their ability to take into account long-term sleep architectural patterns. A long short-term memory (LSTM) network is proposed as a solution to model long-term cardiac sleep architecture information and validated on a comprehensive data set (292 participants, 584 nights, 541.214 annotated 30 s sleep segments) comprising a wide range of ages and pathological profiles, annotated according to the Rechtschaffen and Kales (R&K) annotation standard. It is shown that the model outperforms state-of-the-art approaches which were often limited to non-temporal or short-term recurrent classifiers. The model achieves a Cohen’s k of 0.61 ± 0.15 and accuracy of 77.00 ± 8.90% across the entire database. Further analysis revealed that the performance for individuals aged 50 years and older may decline. These results demonstrate the merit of deep temporal modelling using a diverse data set and advance the state-of-the-art for HRV-based sleep stage classification. Further research is warranted into individuals over the age of 50 as performance tends to worsen in this sub-population.
Original languageEnglish
Article number14149
Number of pages11
JournalScientific Reports
Volume9
DOIs
Publication statusPublished - 2 Oct 2019

Fingerprint

Long-Term Memory
Sleep Stages
Short-Term Memory
Sleep
Polysomnography
Heart Rate
Aptitude
Human Engineering
Brassica
Databases
Costs and Cost Analysis
Research
Population
Datasets

Cite this

@article{ea851afb2aef4f82920462cae6f89b75,
title = "Sleep stage classification from heart-rate variability using long short-term memory neural networks",
abstract = "Automated sleep stage classification using heart rate variability (HRV) may provide an ergonomic and low-cost alternative to gold standard polysomnography, creating possibilities for unobtrusive home-based sleep monitoring. Current methods however are limited in their ability to take into account long-term sleep architectural patterns. A long short-term memory (LSTM) network is proposed as a solution to model long-term cardiac sleep architecture information and validated on a comprehensive data set (292 participants, 584 nights, 541.214 annotated 30 s sleep segments) comprising a wide range of ages and pathological profiles, annotated according to the Rechtschaffen and Kales (R&K) annotation standard. It is shown that the model outperforms state-of-the-art approaches which were often limited to non-temporal or short-term recurrent classifiers. The model achieves a Cohen’s k of 0.61 ± 0.15 and accuracy of 77.00 ± 8.90{\%} across the entire database. Further analysis revealed that the performance for individuals aged 50 years and older may decline. These results demonstrate the merit of deep temporal modelling using a diverse data set and advance the state-of-the-art for HRV-based sleep stage classification. Further research is warranted into individuals over the age of 50 as performance tends to worsen in this sub-population.",
author = "Mustafa Radha and Pedro Fonseca and Arnaud Moreau and Marco Ross and Andreas Cerny and Peter Anderer and Xi Long and Ronald Aarts",
year = "2019",
month = "10",
day = "2",
doi = "10.1038/s41598-019-49703-y",
language = "English",
volume = "9",
journal = "Scientific Reports",
issn = "2045-2322",
publisher = "Nature Publishing Group",

}

Sleep stage classification from heart-rate variability using long short-term memory neural networks. / Radha, Mustafa (Corresponding author); Fonseca, Pedro; Moreau, Arnaud; Ross, Marco; Cerny, Andreas; Anderer, Peter; Long, Xi; Aarts, Ronald.

In: Scientific Reports, Vol. 9, 14149 , 02.10.2019.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Sleep stage classification from heart-rate variability using long short-term memory neural networks

AU - Radha, Mustafa

AU - Fonseca, Pedro

AU - Moreau, Arnaud

AU - Ross, Marco

AU - Cerny, Andreas

AU - Anderer, Peter

AU - Long, Xi

AU - Aarts, Ronald

PY - 2019/10/2

Y1 - 2019/10/2

N2 - Automated sleep stage classification using heart rate variability (HRV) may provide an ergonomic and low-cost alternative to gold standard polysomnography, creating possibilities for unobtrusive home-based sleep monitoring. Current methods however are limited in their ability to take into account long-term sleep architectural patterns. A long short-term memory (LSTM) network is proposed as a solution to model long-term cardiac sleep architecture information and validated on a comprehensive data set (292 participants, 584 nights, 541.214 annotated 30 s sleep segments) comprising a wide range of ages and pathological profiles, annotated according to the Rechtschaffen and Kales (R&K) annotation standard. It is shown that the model outperforms state-of-the-art approaches which were often limited to non-temporal or short-term recurrent classifiers. The model achieves a Cohen’s k of 0.61 ± 0.15 and accuracy of 77.00 ± 8.90% across the entire database. Further analysis revealed that the performance for individuals aged 50 years and older may decline. These results demonstrate the merit of deep temporal modelling using a diverse data set and advance the state-of-the-art for HRV-based sleep stage classification. Further research is warranted into individuals over the age of 50 as performance tends to worsen in this sub-population.

AB - Automated sleep stage classification using heart rate variability (HRV) may provide an ergonomic and low-cost alternative to gold standard polysomnography, creating possibilities for unobtrusive home-based sleep monitoring. Current methods however are limited in their ability to take into account long-term sleep architectural patterns. A long short-term memory (LSTM) network is proposed as a solution to model long-term cardiac sleep architecture information and validated on a comprehensive data set (292 participants, 584 nights, 541.214 annotated 30 s sleep segments) comprising a wide range of ages and pathological profiles, annotated according to the Rechtschaffen and Kales (R&K) annotation standard. It is shown that the model outperforms state-of-the-art approaches which were often limited to non-temporal or short-term recurrent classifiers. The model achieves a Cohen’s k of 0.61 ± 0.15 and accuracy of 77.00 ± 8.90% across the entire database. Further analysis revealed that the performance for individuals aged 50 years and older may decline. These results demonstrate the merit of deep temporal modelling using a diverse data set and advance the state-of-the-art for HRV-based sleep stage classification. Further research is warranted into individuals over the age of 50 as performance tends to worsen in this sub-population.

UR - http://www.scopus.com/inward/record.url?scp=85072925758&partnerID=8YFLogxK

U2 - 10.1038/s41598-019-49703-y

DO - 10.1038/s41598-019-49703-y

M3 - Article

C2 - 31578345

VL - 9

JO - Scientific Reports

JF - Scientific Reports

SN - 2045-2322

M1 - 14149

ER -