Automated interictal EEG spike detection using artificial neural networks

Andrew J. Gabor, Masud Seyal

Research output: Contribution to journalArticle

132 Citations (Scopus)

Abstract

Feed-forward, error-back-propagation artificial neural networks were applied to recognition of epileptiform patterns in the EEG. The inherent network properties of generalization and variability tolerance were effective in identifying wave forms that differed from the training patterns but still maintained 'epileptiform' spatio-temporal characteristics. The certainty of recognition was measured as a continuous function with a range of 0-1. Two levels of certainty (0.825 and 0.900) were used to indicate recognition of spikes and sharp waves (SSW). An average 94.2% (± 7.3) of the SSW were recognized; 20.9% (± 22.9) of all recognized SSW were false-positive recognitions. The time required for pattern recognition was well within the time required for digitizing the analogue data. This study provides evidence that neural network technology is, in principle, an effective pattern recognition strategy for identification of epileptiform transients in the EEG. The analysis is sufficiently rapid to be of potential value as a strategy for data reduction of long recordings stored on bulk media.

Original languageEnglish (US)
Pages (from-to)271-280
Number of pages10
JournalElectroencephalography and Clinical Neurophysiology
Volume83
Issue number5
DOIs
StatePublished - 1992

Fingerprint

Electroencephalography
Technology

Keywords

  • Neural network
  • Spike detection

ASJC Scopus subject areas

  • Clinical Neurology
  • Neuroscience(all)

Cite this

Automated interictal EEG spike detection using artificial neural networks. / Gabor, Andrew J.; Seyal, Masud.

In: Electroencephalography and Clinical Neurophysiology, Vol. 83, No. 5, 1992, p. 271-280.

Research output: Contribution to journalArticle

@article{22fe7d2c217b4ca0947ceb124fa40d9b,
title = "Automated interictal EEG spike detection using artificial neural networks",
abstract = "Feed-forward, error-back-propagation artificial neural networks were applied to recognition of epileptiform patterns in the EEG. The inherent network properties of generalization and variability tolerance were effective in identifying wave forms that differed from the training patterns but still maintained 'epileptiform' spatio-temporal characteristics. The certainty of recognition was measured as a continuous function with a range of 0-1. Two levels of certainty (0.825 and 0.900) were used to indicate recognition of spikes and sharp waves (SSW). An average 94.2{\%} (± 7.3) of the SSW were recognized; 20.9{\%} (± 22.9) of all recognized SSW were false-positive recognitions. The time required for pattern recognition was well within the time required for digitizing the analogue data. This study provides evidence that neural network technology is, in principle, an effective pattern recognition strategy for identification of epileptiform transients in the EEG. The analysis is sufficiently rapid to be of potential value as a strategy for data reduction of long recordings stored on bulk media.",
keywords = "Neural network, Spike detection",
author = "Gabor, {Andrew J.} and Masud Seyal",
year = "1992",
doi = "10.1016/0013-4694(92)90086-W",
language = "English (US)",
volume = "83",
pages = "271--280",
journal = "Electroencephalography and Clinical Neurophysiology",
issn = "0013-4694",
publisher = "Elsevier BV",
number = "5",

}

TY - JOUR

T1 - Automated interictal EEG spike detection using artificial neural networks

AU - Gabor, Andrew J.

AU - Seyal, Masud

PY - 1992

Y1 - 1992

N2 - Feed-forward, error-back-propagation artificial neural networks were applied to recognition of epileptiform patterns in the EEG. The inherent network properties of generalization and variability tolerance were effective in identifying wave forms that differed from the training patterns but still maintained 'epileptiform' spatio-temporal characteristics. The certainty of recognition was measured as a continuous function with a range of 0-1. Two levels of certainty (0.825 and 0.900) were used to indicate recognition of spikes and sharp waves (SSW). An average 94.2% (± 7.3) of the SSW were recognized; 20.9% (± 22.9) of all recognized SSW were false-positive recognitions. The time required for pattern recognition was well within the time required for digitizing the analogue data. This study provides evidence that neural network technology is, in principle, an effective pattern recognition strategy for identification of epileptiform transients in the EEG. The analysis is sufficiently rapid to be of potential value as a strategy for data reduction of long recordings stored on bulk media.

AB - Feed-forward, error-back-propagation artificial neural networks were applied to recognition of epileptiform patterns in the EEG. The inherent network properties of generalization and variability tolerance were effective in identifying wave forms that differed from the training patterns but still maintained 'epileptiform' spatio-temporal characteristics. The certainty of recognition was measured as a continuous function with a range of 0-1. Two levels of certainty (0.825 and 0.900) were used to indicate recognition of spikes and sharp waves (SSW). An average 94.2% (± 7.3) of the SSW were recognized; 20.9% (± 22.9) of all recognized SSW were false-positive recognitions. The time required for pattern recognition was well within the time required for digitizing the analogue data. This study provides evidence that neural network technology is, in principle, an effective pattern recognition strategy for identification of epileptiform transients in the EEG. The analysis is sufficiently rapid to be of potential value as a strategy for data reduction of long recordings stored on bulk media.

KW - Neural network

KW - Spike detection

UR - http://www.scopus.com/inward/record.url?scp=0026675081&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0026675081&partnerID=8YFLogxK

U2 - 10.1016/0013-4694(92)90086-W

DO - 10.1016/0013-4694(92)90086-W

M3 - Article

VL - 83

SP - 271

EP - 280

JO - Electroencephalography and Clinical Neurophysiology

JF - Electroencephalography and Clinical Neurophysiology

SN - 0013-4694

IS - 5

ER -