Feature-Based Detection of the K-Complex Wave in the Human Electroencephalogram Using Neural Networks

Isaac N. Bankman, Vincent G. Sigillito, Robert A Wise, Philip L Smith

Research output: Contribution to journalArticle

Abstract

The main difficulties in reliable automated detection of the K-complex wave in EEG are its close similarity to other waves and the lack of specific characterization criteria. We present a feature-based detection approach using neural networks that provides good agreement with visual K-complex recognition: a sensitivity of 90% is obtained with about 8% false positives. The respective contribution of the features and that of the neural network is demonstrated by comparing the results to those obtained with i) raw EEG data presented to neural networks, and ii) features presented to Fisher's linear discriminant.

Original languageEnglish (US)
Pages (from-to)1305-1310
Number of pages6
JournalIEEE Transactions on Biomedical Engineering
Volume39
Issue number12
DOIs
StatePublished - 1992

Fingerprint

Electroencephalography
Neural networks

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

Feature-Based Detection of the K-Complex Wave in the Human Electroencephalogram Using Neural Networks. / Bankman, Isaac N.; Sigillito, Vincent G.; Wise, Robert A; Smith, Philip L.

In: IEEE Transactions on Biomedical Engineering, Vol. 39, No. 12, 1992, p. 1305-1310.

Research output: Contribution to journalArticle

@article{2a6d0bc4f18c485095f208a350f846d2,
title = "Feature-Based Detection of the K-Complex Wave in the Human Electroencephalogram Using Neural Networks",
abstract = "The main difficulties in reliable automated detection of the K-complex wave in EEG are its close similarity to other waves and the lack of specific characterization criteria. We present a feature-based detection approach using neural networks that provides good agreement with visual K-complex recognition: a sensitivity of 90{\%} is obtained with about 8{\%} false positives. The respective contribution of the features and that of the neural network is demonstrated by comparing the results to those obtained with i) raw EEG data presented to neural networks, and ii) features presented to Fisher's linear discriminant.",
author = "Bankman, {Isaac N.} and Sigillito, {Vincent G.} and Wise, {Robert A} and Smith, {Philip L}",
year = "1992",
doi = "10.1109/10.184707",
language = "English (US)",
volume = "39",
pages = "1305--1310",
journal = "IEEE Transactions on Biomedical Engineering",
issn = "0018-9294",
publisher = "IEEE Computer Society",
number = "12",

}

TY - JOUR

T1 - Feature-Based Detection of the K-Complex Wave in the Human Electroencephalogram Using Neural Networks

AU - Bankman, Isaac N.

AU - Sigillito, Vincent G.

AU - Wise, Robert A

AU - Smith, Philip L

PY - 1992

Y1 - 1992

N2 - The main difficulties in reliable automated detection of the K-complex wave in EEG are its close similarity to other waves and the lack of specific characterization criteria. We present a feature-based detection approach using neural networks that provides good agreement with visual K-complex recognition: a sensitivity of 90% is obtained with about 8% false positives. The respective contribution of the features and that of the neural network is demonstrated by comparing the results to those obtained with i) raw EEG data presented to neural networks, and ii) features presented to Fisher's linear discriminant.

AB - The main difficulties in reliable automated detection of the K-complex wave in EEG are its close similarity to other waves and the lack of specific characterization criteria. We present a feature-based detection approach using neural networks that provides good agreement with visual K-complex recognition: a sensitivity of 90% is obtained with about 8% false positives. The respective contribution of the features and that of the neural network is demonstrated by comparing the results to those obtained with i) raw EEG data presented to neural networks, and ii) features presented to Fisher's linear discriminant.

UR - http://www.scopus.com/inward/record.url?scp=0026969987&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0026969987&partnerID=8YFLogxK

U2 - 10.1109/10.184707

DO - 10.1109/10.184707

M3 - Article

VL - 39

SP - 1305

EP - 1310

JO - IEEE Transactions on Biomedical Engineering

JF - IEEE Transactions on Biomedical Engineering

SN - 0018-9294

IS - 12

ER -