Detection of Postictal Generalized Electroencephalogram Suppression: Random Forest Approach.
EEG
epilepsy
generalized tonic-clonic seizure
postictal generalized EEG suppression
random forest
Journal
JMIR medical informatics
ISSN: 2291-9694
Titre abrégé: JMIR Med Inform
Pays: Canada
ID NLM: 101645109
Informations de publication
Date de publication:
14 Feb 2020
14 Feb 2020
Historique:
received:
15
11
2019
accepted:
29
12
2019
revised:
20
12
2019
entrez:
5
3
2020
pubmed:
5
3
2020
medline:
5
3
2020
Statut:
epublish
Résumé
Sudden unexpected death in epilepsy (SUDEP) is second only to stroke in neurological events resulting in years of potential life lost. Postictal generalized electroencephalogram (EEG) suppression (PGES) is a period of suppressed brain activity often occurring after generalized tonic-clonic seizure, a most significant risk factor for SUDEP. Therefore, PGES has been considered as a potential biomarker for SUDEP risk. Automatic PGES detection tools can address the limitations of labor-intensive, and sometimes inconsistent, visual analysis. A successful approach to automatic PGES detection must overcome computational challenges involved in the detection of subtle amplitude changes in EEG recordings, which may contain physiological and acquisition artifacts. This study aimed to present a random forest approach for automatic PGES detection using multichannel human EEG recordings acquired in epilepsy monitoring units. We used a combination of temporal, frequency, wavelet, and interchannel correlation features derived from EEG signals to train a random forest classifier. We also constructed and applied confidence-based correction rules based on PGES state changes. Motivated by practical utility, we introduced a new, time distance-based evaluation method for assessing the performance of PGES detection algorithms. The time distance-based evaluation showed that our approach achieved a 5-second tolerance-based positive prediction rate of 0.95 for artifact-free signals. For signals with different artifact levels, our prediction rates varied from 0.68 to 0.81. We introduced a feature-based, random forest approach for automatic PGES detection using multichannel EEG recordings. Our approach achieved increasingly better time distance-based performance with reduced signal artifact levels. Further study is needed for PGES detection algorithms to perform well irrespective of the levels of signal artifacts.
Sections du résumé
BACKGROUND
BACKGROUND
Sudden unexpected death in epilepsy (SUDEP) is second only to stroke in neurological events resulting in years of potential life lost. Postictal generalized electroencephalogram (EEG) suppression (PGES) is a period of suppressed brain activity often occurring after generalized tonic-clonic seizure, a most significant risk factor for SUDEP. Therefore, PGES has been considered as a potential biomarker for SUDEP risk. Automatic PGES detection tools can address the limitations of labor-intensive, and sometimes inconsistent, visual analysis. A successful approach to automatic PGES detection must overcome computational challenges involved in the detection of subtle amplitude changes in EEG recordings, which may contain physiological and acquisition artifacts.
OBJECTIVE
OBJECTIVE
This study aimed to present a random forest approach for automatic PGES detection using multichannel human EEG recordings acquired in epilepsy monitoring units.
METHODS
METHODS
We used a combination of temporal, frequency, wavelet, and interchannel correlation features derived from EEG signals to train a random forest classifier. We also constructed and applied confidence-based correction rules based on PGES state changes. Motivated by practical utility, we introduced a new, time distance-based evaluation method for assessing the performance of PGES detection algorithms.
RESULTS
RESULTS
The time distance-based evaluation showed that our approach achieved a 5-second tolerance-based positive prediction rate of 0.95 for artifact-free signals. For signals with different artifact levels, our prediction rates varied from 0.68 to 0.81.
CONCLUSIONS
CONCLUSIONS
We introduced a feature-based, random forest approach for automatic PGES detection using multichannel EEG recordings. Our approach achieved increasingly better time distance-based performance with reduced signal artifact levels. Further study is needed for PGES detection algorithms to perform well irrespective of the levels of signal artifacts.
Identifiants
pubmed: 32130173
pii: v8i2e17061
doi: 10.2196/17061
pmc: PMC7055778
doi:
Types de publication
Journal Article
Langues
eng
Pagination
e17061Subventions
Organisme : NCATS NIH HHS
ID : UL1 TR003167
Pays : United States
Informations de copyright
©Xiaojin Li, Shiqiang Tao, Shirin Jamal-Omidi, Yan Huang, Samden D Lhatoo, Guo-Qiang Zhang, Licong Cui. Originally published in JMIR Medical Informatics (http://medinform.jmir.org), 14.02.2020.
Références
Comput Biol Med. 2011 Jun;41(6):380-9
pubmed: 21497802
Ann Indian Acad Neurol. 2014 Mar;17(Suppl 1):S40-4
pubmed: 24791088
Electroencephalogr Clin Neurophysiol. 1970 Sep;29(3):306-10
pubmed: 4195653
Front Neurol. 2018 Aug 09;9:639
pubmed: 30140254
Biomark Med. 2011 Oct;5(5):557-66
pubmed: 22003904
IEEE J Biomed Health Inform. 2018 Mar;22(2):375-385
pubmed: 28222004
Epilepsia. 2015 Nov;56(11):1700-6
pubmed: 26494436
Front Neurol. 2019 Mar 01;10:166
pubmed: 30890997
IEEE Trans Neural Syst Rehabil Eng. 2004 Dec;12(4):406-15
pubmed: 15614996
Cochrane Database Syst Rev. 2016 Jul 19;7:CD011792
pubmed: 27434597
IEEE Trans Biomed Eng. 2017 Sep;64(9):2003-2015
pubmed: 28092514
J Neural Eng. 2007 Jun;4(2):120-9
pubmed: 17409486
Epilepsy Behav. 2016 May;58:22-5
pubmed: 26994879
Epilepsia. 2016 Mar;57(3):412-7
pubmed: 26763069
Epilepsy Curr. 2001 Sep;1(1):21-23
pubmed: 15309034
IEEE Trans Biomed Eng. 2006 Mar;53(3):485-96
pubmed: 16532775
Conf Proc IEEE Eng Med Biol Soc. 2007;2007:2520-3
pubmed: 18002507
Epilepsia. 2014 Oct;55(10):1479-85
pubmed: 24903551
Sci Rep. 2019 May 10;9(1):7243
pubmed: 31076609
Ann Neurol. 2010 Dec;68(6):787-96
pubmed: 20882604
Neurology. 2015 Nov 3;85(18):1598-603
pubmed: 26333799
Comput Methods Programs Biomed. 2012 Oct;108(1):10-9
pubmed: 22178068
Epilepsia. 2016 Jul;57(7):1161-8
pubmed: 27221596
Seizure. 2018 Oct;61:135-138
pubmed: 30142618
Comput Biol Med. 2012 Dec;42(12):1186-95
pubmed: 23102750
P T. 2010 Jul;35(7):392-415
pubmed: 20689626
J Neurol Neurosurg Psychiatry. 2005 Jun;76 Suppl 2:ii2-7
pubmed: 15961864
IEEE Trans Biomed Eng. 2018 Feb;65(2):371-377
pubmed: 29346105
Front Neurol. 2018 Sep 26;9:793
pubmed: 30319527
Electroencephalogr Clin Neurophysiol. 1983 Apr;55(4):468-84
pubmed: 6187540
Neurotherapeutics. 2014 Apr;11(2):334-46
pubmed: 24519238
Comput Biol Med. 2017 Sep 1;88:132-141
pubmed: 28719805