Deep residual networks for automatic sleep stage classification of raw polysomnographic waveforms

Alexander N. Olesen; Poul Jennum; Paul Peppard; Emmanuel Mignot; Helge B.D. Sorensen

doi:10.1109/EMBC.2018.8513080

Deep residual networks for automatic sleep stage classification of raw polysomnographic waveforms

Alexander N. Olesen^*, Poul Jennum, Paul Peppard, Emmanuel Mignot, Helge B.D. Sorensen

^*Corresponding author af dette arbejde

Institut for Klinisk Medicin

10 Citationer (Scopus)

Abstract

We have developed an automatic sleep stage classification algorithm based on deep residual neural networks and raw polysomnogram signals. Briefly, the raw data is passed through 50 convolutional layers before subsequent classification into one of five sleep stages. Three model configurations were trained on 1850 polysomnogram recordings and subsequently tested on 230 independent recordings. Our best performing model yielded an accuracy of 84.1% and a Cohen's kappa of 0.746, improving on previous reported results by other groups also using only raw polysomnogram data. Most errors were made on non-REM stage 1 and 3 decisions, errors likely resulting from the definition of these stages. Further testing on independent cohorts is needed to verify performance for clinical use.

Originalsprog	Engelsk
Titel	40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018
Antal sider	4
Forlag	IEEE
Publikationsdato	2018
Sider	3713-3716
Artikelnummer	8513080
ISBN (Elektronisk)	978-1-5386-3646-6
DOI	https://doi.org/10.1109/EMBC.2018.8513080
Status	Udgivet - 2018
Begivenhed	40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018 - Honolulu, USA Varighed: 18 jul. 2018 → 21 jul. 2018

Konference

Konference	40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018
Land/Område	USA
By	Honolulu
Periode	18/07/2018 → 21/07/2018

Navn	Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS
Vol/bind	2018-July
ISSN	1557-170X

Adgang til dokumentet

10.1109/EMBC.2018.8513080

Citationsformater

Deep residual networks for automatic sleep stage classification of raw polysomnographic waveforms. / Olesen, Alexander N.; Jennum, Poul; Peppard, Paul et al.
40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018. IEEE, 2018. s. 3713-3716 8513080 (Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, Bind 2018-July).

Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › peer review

Olesen, AN, Jennum, P, Peppard, P, Mignot, E & Sorensen, HBD 2018, Deep residual networks for automatic sleep stage classification of raw polysomnographic waveforms. i 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018., 8513080, IEEE, Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, bind 2018-July, s. 3713-3716, 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018, Honolulu, USA, 18/07/2018. https://doi.org/10.1109/EMBC.2018.8513080

Olesen AN, Jennum P, Peppard P, Mignot E, Sorensen HBD. Deep residual networks for automatic sleep stage classification of raw polysomnographic waveforms. I 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018. IEEE. 2018. s. 3713-3716. 8513080. (Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, Bind 2018-July). doi: 10.1109/EMBC.2018.8513080

Olesen, Alexander N. ; Jennum, Poul ; Peppard, Paul et al. / Deep residual networks for automatic sleep stage classification of raw polysomnographic waveforms. 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018. IEEE, 2018. s. 3713-3716 (Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, Bind 2018-July).

@inproceedings{238f0894ab6f46cf96a7b7cf8b2b6af6,

title = "Deep residual networks for automatic sleep stage classification of raw polysomnographic waveforms",

abstract = "We have developed an automatic sleep stage classification algorithm based on deep residual neural networks and raw polysomnogram signals. Briefly, the raw data is passed through 50 convolutional layers before subsequent classification into one of five sleep stages. Three model configurations were trained on 1850 polysomnogram recordings and subsequently tested on 230 independent recordings. Our best performing model yielded an accuracy of 84.1% and a Cohen's kappa of 0.746, improving on previous reported results by other groups also using only raw polysomnogram data. Most errors were made on non-REM stage 1 and 3 decisions, errors likely resulting from the definition of these stages. Further testing on independent cohorts is needed to verify performance for clinical use.",

author = "Olesen, {Alexander N.} and Poul Jennum and Paul Peppard and Emmanuel Mignot and Sorensen, {Helge B.D.}",

year = "2018",

doi = "10.1109/EMBC.2018.8513080",

language = "English",

series = "Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS",

publisher = "IEEE",

pages = "3713--3716",

booktitle = "40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018",

note = "40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018 ; Conference date: 18-07-2018 Through 21-07-2018",

}

TY - GEN

T1 - Deep residual networks for automatic sleep stage classification of raw polysomnographic waveforms

AU - Olesen, Alexander N.

AU - Jennum, Poul

AU - Peppard, Paul

AU - Mignot, Emmanuel

AU - Sorensen, Helge B.D.

PY - 2018

Y1 - 2018

N2 - We have developed an automatic sleep stage classification algorithm based on deep residual neural networks and raw polysomnogram signals. Briefly, the raw data is passed through 50 convolutional layers before subsequent classification into one of five sleep stages. Three model configurations were trained on 1850 polysomnogram recordings and subsequently tested on 230 independent recordings. Our best performing model yielded an accuracy of 84.1% and a Cohen's kappa of 0.746, improving on previous reported results by other groups also using only raw polysomnogram data. Most errors were made on non-REM stage 1 and 3 decisions, errors likely resulting from the definition of these stages. Further testing on independent cohorts is needed to verify performance for clinical use.

AB - We have developed an automatic sleep stage classification algorithm based on deep residual neural networks and raw polysomnogram signals. Briefly, the raw data is passed through 50 convolutional layers before subsequent classification into one of five sleep stages. Three model configurations were trained on 1850 polysomnogram recordings and subsequently tested on 230 independent recordings. Our best performing model yielded an accuracy of 84.1% and a Cohen's kappa of 0.746, improving on previous reported results by other groups also using only raw polysomnogram data. Most errors were made on non-REM stage 1 and 3 decisions, errors likely resulting from the definition of these stages. Further testing on independent cohorts is needed to verify performance for clinical use.

U2 - 10.1109/EMBC.2018.8513080

DO - 10.1109/EMBC.2018.8513080

M3 - Article in proceedings

C2 - 30440296

AN - SCOPUS:85056649177

T3 - Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS

SP - 3713

EP - 3716

BT - 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018

PB - IEEE

T2 - 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018

Y2 - 18 July 2018 through 21 July 2018

ER -

Deep residual networks for automatic sleep stage classification of raw polysomnographic waveforms

Abstract

Konference

Adgang til dokumentet

Fingeraftryk

Citationsformater