A sleep spindle detection algorithm that emulates human expert spindle scoring.
Detector
Electroencephalography (EEG)
Polysomnography (PSG)
Sigma
Sleep
Sleep spindle
Journal
Journal of neuroscience methods
ISSN: 1872-678X
Titre abrégé: J Neurosci Methods
Pays: Netherlands
ID NLM: 7905558
Informations de publication
Date de publication:
15 03 2019
15 03 2019
Historique:
received:
11
04
2018
revised:
10
08
2018
accepted:
10
08
2018
pubmed:
15
8
2018
medline:
29
7
2020
entrez:
15
8
2018
Statut:
ppublish
Résumé
Sleep spindles are a marker of stage 2 NREM sleep that are linked to learning & memory and are altered by many neurological diseases. Although visual inspection of the EEG is considered the gold standard for spindle detection, it is time-consuming, costly and can introduce inter/ra-scorer bias. Our goal was to develop a simple and efficient sleep-spindle detector (algorithm #7, or 'A7') that emulates human scoring. 'A7' runs on a single EEG channel and relies on four parameters: the absolute sigma power, relative sigma power, and correlation/covariance of the sigma band-passed signal to the original EEG signal. To test the performance of the detector, we compared it against a gold standard spindle dataset derived from the consensus of a group of human experts. The by-event performance of the 'A7' spindle detector was 74% precision, 68% recall (sensitivity), and an F1-score of 0.70. This performance was equivalent to an individual human expert (average F1-score = 0.67). The F1-score of 'A7' was 0.17 points higher than other spindle detectors tested. Existing detectors have a tendency to find large numbers of false positives compared to human scorers. On a by-subject basis, the spindle density estimates produced by A7 were well correlated with human experts (r The 'A7' detector is a sensitive and precise tool designed to emulate human spindle scoring by minimizing the number of 'hidden spindles' detected. We provide an open-source implementation of this detector for further use and testing.
Sections du résumé
BACKGROUND
Sleep spindles are a marker of stage 2 NREM sleep that are linked to learning & memory and are altered by many neurological diseases. Although visual inspection of the EEG is considered the gold standard for spindle detection, it is time-consuming, costly and can introduce inter/ra-scorer bias.
NEW METHOD
Our goal was to develop a simple and efficient sleep-spindle detector (algorithm #7, or 'A7') that emulates human scoring. 'A7' runs on a single EEG channel and relies on four parameters: the absolute sigma power, relative sigma power, and correlation/covariance of the sigma band-passed signal to the original EEG signal. To test the performance of the detector, we compared it against a gold standard spindle dataset derived from the consensus of a group of human experts.
RESULTS
The by-event performance of the 'A7' spindle detector was 74% precision, 68% recall (sensitivity), and an F1-score of 0.70. This performance was equivalent to an individual human expert (average F1-score = 0.67).
COMPARISON WITH EXISTING METHOD(S)
The F1-score of 'A7' was 0.17 points higher than other spindle detectors tested. Existing detectors have a tendency to find large numbers of false positives compared to human scorers. On a by-subject basis, the spindle density estimates produced by A7 were well correlated with human experts (r
CONCLUSIONS
The 'A7' detector is a sensitive and precise tool designed to emulate human spindle scoring by minimizing the number of 'hidden spindles' detected. We provide an open-source implementation of this detector for further use and testing.
Identifiants
pubmed: 30107208
pii: S0165-0270(18)30250-4
doi: 10.1016/j.jneumeth.2018.08.014
pmc: PMC6415669
mid: NIHMS1510485
pii:
doi:
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
3-11Subventions
Organisme : NHLBI NIH HHS
ID : R01 HL062252
Pays : United States
Organisme : NCRR NIH HHS
ID : UL1 RR025011
Pays : United States
Organisme : NCATS NIH HHS
ID : UL1 TR002373
Pays : United States
Organisme : CIHR
ID : OOGP 313177
Pays : Canada
Informations de copyright
Copyright © 2018 Elsevier B.V. All rights reserved.
Références
Clin Neurophysiol. 2001 Aug;112(8):1540-52
pubmed: 11459695
J Neurosci. 2002 Dec 15;22(24):10941-7
pubmed: 12486189
Sleep Med Rev. 2003 Oct;7(5):423-40
pubmed: 14573378
J Sleep Res. 2004 Mar;13(1):63-9
pubmed: 14996037
Clin Neurophysiol. 2004 Apr;115(4):938-45
pubmed: 15003776
J Psychosom Res. 2004 May;56(5):487-96
pubmed: 15172204
Neuropsychobiology. 2004;50(2):147-52
pubmed: 15292669
Brain. 2005 May;128(Pt 5):1049-61
pubmed: 15705609
Am J Psychiatry. 2007 Mar;164(3):483-92
pubmed: 17329474
Artif Intell Med. 2007 Jul;40(3):157-70
pubmed: 17555950
J Clin Sleep Med. 2007 Mar 15;3(2):121-31
pubmed: 17557422
J Psychosom Res. 2009 Jan;66(1):59-65
pubmed: 19073295
Ann N Y Acad Sci. 2009 Mar;1156:168-97
pubmed: 19338508
Nat Rev Neurosci. 2010 Feb;11(2):114-26
pubmed: 20046194
Neurosci Biobehav Rev. 2011 Apr;35(5):1154-65
pubmed: 21167865
Biol Psychiatry. 2012 Jan 15;71(2):154-61
pubmed: 21967958
Neurobiol Aging. 2013 Feb;34(2):468-76
pubmed: 22809452
Am J Epidemiol. 2013 May 1;177(9):1006-14
pubmed: 23589584
Clin Neurophysiol. 2014 Mar;125(3):512-9
pubmed: 24125856
Nat Methods. 2014 Apr;11(4):385-92
pubmed: 24562424
Sleep Disord. 2014;2014:271802
pubmed: 24800086
Clin Neurophysiol. 2015 Aug;126(8):1548-56
pubmed: 25434753
Front Hum Neurosci. 2015 Feb 10;9:68
pubmed: 25713529
Int J Psychophysiol. 2015 Jul;97(1):58-65
pubmed: 25958790
Neural Plast. 2016;2016:7328725
pubmed: 27034850
Neural Plast. 2016;2016:4724792
pubmed: 27478646
Front Hum Neurosci. 2017 Jan 18;10:672
pubmed: 28149273
Sleep Med. 2017 Jun;34:40-49
pubmed: 28522097
Electroencephalogr Clin Neurophysiol. 1976 Jun;40(6):666-70
pubmed: 57053
Electroencephalogr Clin Neurophysiol. 1997 Nov;103(5):535-42
pubmed: 9402884