Wavelet preprocessing for automated neural network detection of EEG spikes

T. Kalayci, Ozcan Ozdamar

Research output: Contribution to journalArticle

184 Citations (Scopus)

Abstract

The feasibility of using a wavelet transform (WT) as a preprocessor for an automated neural network (ANN)-based EEG spike detection system was confirmed. The study aimed at decreasing the input size to the ANN detector, without decreasing the information content of the signal and degrading the detection performance. Because routine clinical EEG requires recordings for many channels, input size becomes a critical design parameter for real-time multichannel spike detection systems. For a sliding window of 20 points, more than 600 input lines will be necessary for a 32-channel system, which is not easily manageable with current ANN technology.

Original languageEnglish
Pages (from-to)160-166
Number of pages7
JournalIEEE Engineering in Medicine and Biology Magazine
Volume14
Issue number2
DOIs
StatePublished - Mar 1 1995

Fingerprint

Electroencephalography
Neural networks
Wavelet Analysis
Technology
Wavelet transforms
Detectors

ASJC Scopus subject areas

  • Biomedical Engineering
  • Health Informatics
  • Health Information Management

Cite this

Wavelet preprocessing for automated neural network detection of EEG spikes. / Kalayci, T.; Ozdamar, Ozcan.

In: IEEE Engineering in Medicine and Biology Magazine, Vol. 14, No. 2, 01.03.1995, p. 160-166.

Research output: Contribution to journalArticle

@article{e1de0898c30e43b1b1b61f8ee43fa03d,
title = "Wavelet preprocessing for automated neural network detection of EEG spikes",
abstract = "The feasibility of using a wavelet transform (WT) as a preprocessor for an automated neural network (ANN)-based EEG spike detection system was confirmed. The study aimed at decreasing the input size to the ANN detector, without decreasing the information content of the signal and degrading the detection performance. Because routine clinical EEG requires recordings for many channels, input size becomes a critical design parameter for real-time multichannel spike detection systems. For a sliding window of 20 points, more than 600 input lines will be necessary for a 32-channel system, which is not easily manageable with current ANN technology.",
author = "T. Kalayci and Ozcan Ozdamar",
year = "1995",
month = "3",
day = "1",
doi = "10.1109/51.376754",
language = "English",
volume = "14",
pages = "160--166",
journal = "IEEE Pulse",
issn = "2154-2287",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

TY - JOUR

T1 - Wavelet preprocessing for automated neural network detection of EEG spikes

AU - Kalayci, T.

AU - Ozdamar, Ozcan

PY - 1995/3/1

Y1 - 1995/3/1

N2 - The feasibility of using a wavelet transform (WT) as a preprocessor for an automated neural network (ANN)-based EEG spike detection system was confirmed. The study aimed at decreasing the input size to the ANN detector, without decreasing the information content of the signal and degrading the detection performance. Because routine clinical EEG requires recordings for many channels, input size becomes a critical design parameter for real-time multichannel spike detection systems. For a sliding window of 20 points, more than 600 input lines will be necessary for a 32-channel system, which is not easily manageable with current ANN technology.

AB - The feasibility of using a wavelet transform (WT) as a preprocessor for an automated neural network (ANN)-based EEG spike detection system was confirmed. The study aimed at decreasing the input size to the ANN detector, without decreasing the information content of the signal and degrading the detection performance. Because routine clinical EEG requires recordings for many channels, input size becomes a critical design parameter for real-time multichannel spike detection systems. For a sliding window of 20 points, more than 600 input lines will be necessary for a 32-channel system, which is not easily manageable with current ANN technology.

UR - http://www.scopus.com/inward/record.url?scp=0029256646&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029256646&partnerID=8YFLogxK

U2 - 10.1109/51.376754

DO - 10.1109/51.376754

M3 - Article

AN - SCOPUS:0029256646

VL - 14

SP - 160

EP - 166

JO - IEEE Pulse

JF - IEEE Pulse

SN - 2154-2287

IS - 2

ER -