Time-Varying EEG Correlations Improve Automated Neonatal Seizure Detection

Show full item record



Permalink

http://hdl.handle.net/10138/311546

Citation

Tapani , K T , Vanhatalo , S & Stevenson , N J 2019 , ' Time-Varying EEG Correlations Improve Automated Neonatal Seizure Detection ' , International Journal of Neural Systems , vol. 29 , no. 4 , 1850030 . https://doi.org/10.1142/S0129065718500302

Title: Time-Varying EEG Correlations Improve Automated Neonatal Seizure Detection
Author: Tapani, Karoliina T.; Vanhatalo, Sampsa; Stevenson, Nathan J.
Contributor organization: Department of Neurosciences
HYKS erva
Kliinisen neurofysiologian yksikkö
Clinicum
Children's Hospital
Department of Diagnostics and Therapeutics
HUS Children and Adolescents
HUS Medical Imaging Center
HUSLAB
Faculty of Medicine
Date: 2019-05
Language: eng
Number of pages: 15
Belongs to series: International Journal of Neural Systems
ISSN: 0129-0657
DOI: https://doi.org/10.1142/S0129065718500302
URI: http://hdl.handle.net/10138/311546
Abstract: The aim of this study was to develop methods for detecting the nonstationary periodic characteristics of neonatal electroencephalographic (EEG) seizures by adapting estimates of the correlation both in the time (spike correlation; SC) and time-frequency domain (time-frequency correlation; TFC). These measures were incorporated into a seizure detection algorithm (SDA) based on a support vector machine to detect periods of seizure and nonseizure. The performance of these nonstationary correlation measures was evaluated using EEG recordings from 79 term neonates annotated by three human experts. The proposed measures were highly discriminative for seizure detection (median AUC(SC): 0.933 IQR: 0.821-0.975, median AUC(TFC): 0.883 IQR: 0.707-0.931). The resultant SDA applied to multi-channel recordings had a median AUC of 0.988 (IQR: 0.931-0.998) when compared to consensus annotations, outperformed two state-of-the-art SDAs (p <0.001) and was noninferior to the human expert for 73/79 of neonates.
Subject: Electroencephalography
support vector machines
time-frequency distributions
neonatal seizure detection
nonstationary signal processing
STATISTICAL TURING TEST
NEWBORN EEG
BRAIN-INJURY
FREQUENCY
MODEL
ALGORITHM
AGREEMENT
INFANTS
SYSTEM
SERIES
3112 Neurosciences
3124 Neurology and psychiatry
113 Computer and information sciences
Peer reviewed: Yes
Rights: unspecified
Usage restriction: openAccess
Self-archived version: acceptedVersion


Files in this item

Total number of downloads: Loading...

Files Size Format View
manuscript.pdf 572.5Kb PDF View/Open

This item appears in the following Collection(s)

Show full item record