Cargando…

Deep learning enables sleep staging from photoplethysmogram for patients with suspected sleep apnea

STUDY OBJECTIVES: Accurate identification of sleep stages is essential in the diagnosis of sleep disorders (e.g. obstructive sleep apnea [OSA]) but relies on labor-intensive electroencephalogram (EEG)-based manual scoring. Furthermore, long-term assessment of sleep relies on actigraphy differentiati...

Descripción completa

Detalles Bibliográficos
Autores principales: Korkalainen, Henri, Aakko, Juhani, Duce, Brett, Kainulainen, Samu, Leino, Akseli, Nikkonen, Sami, Afara, Isaac O, Myllymaa, Sami, Töyräs, Juha, Leppänen, Timo
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Oxford University Press 2020
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7658638/
https://www.ncbi.nlm.nih.gov/pubmed/32436942
http://dx.doi.org/10.1093/sleep/zsaa098
_version_ 1783608715777146880
author Korkalainen, Henri
Aakko, Juhani
Duce, Brett
Kainulainen, Samu
Leino, Akseli
Nikkonen, Sami
Afara, Isaac O
Myllymaa, Sami
Töyräs, Juha
Leppänen, Timo
author_facet Korkalainen, Henri
Aakko, Juhani
Duce, Brett
Kainulainen, Samu
Leino, Akseli
Nikkonen, Sami
Afara, Isaac O
Myllymaa, Sami
Töyräs, Juha
Leppänen, Timo
author_sort Korkalainen, Henri
collection PubMed
description STUDY OBJECTIVES: Accurate identification of sleep stages is essential in the diagnosis of sleep disorders (e.g. obstructive sleep apnea [OSA]) but relies on labor-intensive electroencephalogram (EEG)-based manual scoring. Furthermore, long-term assessment of sleep relies on actigraphy differentiating only between wake and sleep periods without identifying specific sleep stages and having low reliability in identifying wake periods after sleep onset. To address these issues, we aimed to develop an automatic method for identifying the sleep stages from the photoplethysmogram (PPG) signal obtained with a simple finger pulse oximeter. METHODS: PPG signals from the diagnostic polysomnographies of susptected OSA patients (n = 894) were utilized to develop a combined convolutional and recurrent neural network. The deep learning model was trained individually for three-stage (wake/NREM/REM), four-stage (wake/N1+N2/N3/REM), and five-stage (wake/N1/N2/N3/REM) classification of sleep. RESULTS: The three-stage model achieved an epoch-by-epoch accuracy of 80.1% with Cohen’s κ of 0.65. The four- and five-stage models achieved 68.5% (κ = 0.54), and 64.1% (κ = 0.51) accuracies, respectively. With the five-stage model, the total sleep time was underestimated with a mean bias error (SD) of of 7.5 (55.2) minutes. CONCLUSION: The PPG-based deep learning model enabled accurate estimation of sleep time and differentiation between sleep stages with a moderate agreement to manual EEG-based scoring. As PPG is already included in ambulatory polygraphic recordings, applying the PPG-based sleep staging could improve their diagnostic value by enabling simple, low-cost, and reliable monitoring of sleep and help assess otherwise overlooked conditions such as REM-related OSA.
format Online
Article
Text
id pubmed-7658638
institution National Center for Biotechnology Information
language English
publishDate 2020
publisher Oxford University Press
record_format MEDLINE/PubMed
spelling pubmed-76586382020-11-18 Deep learning enables sleep staging from photoplethysmogram for patients with suspected sleep apnea Korkalainen, Henri Aakko, Juhani Duce, Brett Kainulainen, Samu Leino, Akseli Nikkonen, Sami Afara, Isaac O Myllymaa, Sami Töyräs, Juha Leppänen, Timo Sleep Big Data Approaches to Sleep and Circadian Science STUDY OBJECTIVES: Accurate identification of sleep stages is essential in the diagnosis of sleep disorders (e.g. obstructive sleep apnea [OSA]) but relies on labor-intensive electroencephalogram (EEG)-based manual scoring. Furthermore, long-term assessment of sleep relies on actigraphy differentiating only between wake and sleep periods without identifying specific sleep stages and having low reliability in identifying wake periods after sleep onset. To address these issues, we aimed to develop an automatic method for identifying the sleep stages from the photoplethysmogram (PPG) signal obtained with a simple finger pulse oximeter. METHODS: PPG signals from the diagnostic polysomnographies of susptected OSA patients (n = 894) were utilized to develop a combined convolutional and recurrent neural network. The deep learning model was trained individually for three-stage (wake/NREM/REM), four-stage (wake/N1+N2/N3/REM), and five-stage (wake/N1/N2/N3/REM) classification of sleep. RESULTS: The three-stage model achieved an epoch-by-epoch accuracy of 80.1% with Cohen’s κ of 0.65. The four- and five-stage models achieved 68.5% (κ = 0.54), and 64.1% (κ = 0.51) accuracies, respectively. With the five-stage model, the total sleep time was underestimated with a mean bias error (SD) of of 7.5 (55.2) minutes. CONCLUSION: The PPG-based deep learning model enabled accurate estimation of sleep time and differentiation between sleep stages with a moderate agreement to manual EEG-based scoring. As PPG is already included in ambulatory polygraphic recordings, applying the PPG-based sleep staging could improve their diagnostic value by enabling simple, low-cost, and reliable monitoring of sleep and help assess otherwise overlooked conditions such as REM-related OSA. Oxford University Press 2020-05-21 /pmc/articles/PMC7658638/ /pubmed/32436942 http://dx.doi.org/10.1093/sleep/zsaa098 Text en © Sleep Research Society 2020. Published by Oxford University Press on behalf of the Sleep Research Society. http://creativecommons.org/licenses/by-nc/4.0/ This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com
spellingShingle Big Data Approaches to Sleep and Circadian Science
Korkalainen, Henri
Aakko, Juhani
Duce, Brett
Kainulainen, Samu
Leino, Akseli
Nikkonen, Sami
Afara, Isaac O
Myllymaa, Sami
Töyräs, Juha
Leppänen, Timo
Deep learning enables sleep staging from photoplethysmogram for patients with suspected sleep apnea
title Deep learning enables sleep staging from photoplethysmogram for patients with suspected sleep apnea
title_full Deep learning enables sleep staging from photoplethysmogram for patients with suspected sleep apnea
title_fullStr Deep learning enables sleep staging from photoplethysmogram for patients with suspected sleep apnea
title_full_unstemmed Deep learning enables sleep staging from photoplethysmogram for patients with suspected sleep apnea
title_short Deep learning enables sleep staging from photoplethysmogram for patients with suspected sleep apnea
title_sort deep learning enables sleep staging from photoplethysmogram for patients with suspected sleep apnea
topic Big Data Approaches to Sleep and Circadian Science
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7658638/
https://www.ncbi.nlm.nih.gov/pubmed/32436942
http://dx.doi.org/10.1093/sleep/zsaa098
work_keys_str_mv AT korkalainenhenri deeplearningenablessleepstagingfromphotoplethysmogramforpatientswithsuspectedsleepapnea
AT aakkojuhani deeplearningenablessleepstagingfromphotoplethysmogramforpatientswithsuspectedsleepapnea
AT ducebrett deeplearningenablessleepstagingfromphotoplethysmogramforpatientswithsuspectedsleepapnea
AT kainulainensamu deeplearningenablessleepstagingfromphotoplethysmogramforpatientswithsuspectedsleepapnea
AT leinoakseli deeplearningenablessleepstagingfromphotoplethysmogramforpatientswithsuspectedsleepapnea
AT nikkonensami deeplearningenablessleepstagingfromphotoplethysmogramforpatientswithsuspectedsleepapnea
AT afaraisaaco deeplearningenablessleepstagingfromphotoplethysmogramforpatientswithsuspectedsleepapnea
AT myllymaasami deeplearningenablessleepstagingfromphotoplethysmogramforpatientswithsuspectedsleepapnea
AT toyrasjuha deeplearningenablessleepstagingfromphotoplethysmogramforpatientswithsuspectedsleepapnea
AT leppanentimo deeplearningenablessleepstagingfromphotoplethysmogramforpatientswithsuspectedsleepapnea