Common sleep data pipeline for combined data sets

Jesper Strøm; Andreas Larsen Engholm; Kristian Peter Lorenzen; Kaare B. Mikkelsen

doi:10.1371/journal.pone.0307202

Common sleep data pipeline for combined data sets

Jesper Strøm^*, Andreas Larsen Engholm, Kristian Peter Lorenzen, Kaare B. Mikkelsen

^*Corresponding author for this work

Department of Electrical and Computer Engineering - Biomedical Engineering

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

Abstract

Over the past few years, sleep research has shown impressive performance of deep neural networks in the area of automatic sleep-staging. Recent studies have demonstrated the necessity of combining multiple data sets to obtain sufficiently generalizing results. However, working with large amounts of sleep data can be challenging, both from a hardware perspective and because of the different preprocessing steps necessary for distinct data sources. Here we review the possible obstacles and present an open-source pipeline for automatic data loading. Our solution includes both a standardized data store as well as a ‘data serving’ portion which can be used to train neural networks on the standardized data, allowing for different configuration options for different studies and machine learning designs. The pipeline, including implementation, is made public to ensure better and more reproducible sleep research.

Original language	English
Article number	e0307202
Journal	PLOS ONE
Volume	19
Issue	8
ISSN	1932-6203
DOIs	https://doi.org/10.1371/journal.pone.0307202
Publication status	Published - Aug 2024

Access to Document

10.1371/journal.pone.0307202Licence: CC BY

Cite this

@article{fd08670dde574bf892fed185640ca548,

title = "Common sleep data pipeline for combined data sets",

abstract = "Over the past few years, sleep research has shown impressive performance of deep neural networks in the area of automatic sleep-staging. Recent studies have demonstrated the necessity of combining multiple data sets to obtain sufficiently generalizing results. However, working with large amounts of sleep data can be challenging, both from a hardware perspective and because of the different preprocessing steps necessary for distinct data sources. Here we review the possible obstacles and present an open-source pipeline for automatic data loading. Our solution includes both a standardized data store as well as a {\textquoteleft}data serving{\textquoteright} portion which can be used to train neural networks on the standardized data, allowing for different configuration options for different studies and machine learning designs. The pipeline, including implementation, is made public to ensure better and more reproducible sleep research.",

author = "Jesper Str{\o}m and Engholm, {Andreas Larsen} and Lorenzen, {Kristian Peter} and Mikkelsen, {Kaare B.}",

note = "Publisher Copyright: Copyright: {\textcopyright} 2024 Str{\o}m et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.",

year = "2024",

month = aug,

doi = "10.1371/journal.pone.0307202",

language = "English",

volume = "19",

journal = "PLOS ONE",

issn = "1932-6203",

publisher = "public library of science",

number = "8",

}

TY - JOUR

T1 - Common sleep data pipeline for combined data sets

AU - Strøm, Jesper

AU - Engholm, Andreas Larsen

AU - Lorenzen, Kristian Peter

AU - Mikkelsen, Kaare B.

N1 - Publisher Copyright: Copyright: © 2024 Strøm et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

PY - 2024/8

Y1 - 2024/8

N2 - Over the past few years, sleep research has shown impressive performance of deep neural networks in the area of automatic sleep-staging. Recent studies have demonstrated the necessity of combining multiple data sets to obtain sufficiently generalizing results. However, working with large amounts of sleep data can be challenging, both from a hardware perspective and because of the different preprocessing steps necessary for distinct data sources. Here we review the possible obstacles and present an open-source pipeline for automatic data loading. Our solution includes both a standardized data store as well as a ‘data serving’ portion which can be used to train neural networks on the standardized data, allowing for different configuration options for different studies and machine learning designs. The pipeline, including implementation, is made public to ensure better and more reproducible sleep research.

AB - Over the past few years, sleep research has shown impressive performance of deep neural networks in the area of automatic sleep-staging. Recent studies have demonstrated the necessity of combining multiple data sets to obtain sufficiently generalizing results. However, working with large amounts of sleep data can be challenging, both from a hardware perspective and because of the different preprocessing steps necessary for distinct data sources. Here we review the possible obstacles and present an open-source pipeline for automatic data loading. Our solution includes both a standardized data store as well as a ‘data serving’ portion which can be used to train neural networks on the standardized data, allowing for different configuration options for different studies and machine learning designs. The pipeline, including implementation, is made public to ensure better and more reproducible sleep research.

UR - http://www.scopus.com/inward/record.url?scp=85200828292&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0307202

DO - 10.1371/journal.pone.0307202

M3 - Journal article

C2 - 39106236

AN - SCOPUS:85200828292

SN - 1932-6203

VL - 19

JO - PLOS ONE

JF - PLOS ONE

IS - 8

M1 - e0307202

ER -

Common sleep data pipeline for combined data sets

Abstract

Access to Document

Other files and links

Fingerprint

Cite this