Newborn EEG seizure detection based on interspike space distribution in the time-frequency domain

H. Hassanpour, M. Mesbah

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

This paper presents a new time-frequency based EEG seizure detection method. This method uses the distribution of interspike intervals as a criterion for discriminating between seizure and nonseizure activities. To detect spikes in the EEG, the signal is mapped into the time-frequency domain. The high instantaneous energy of spikes is reflected as a localized energy in time-frequency domain. Histogram of successive spikes intervals is then used as a feature for seizure detection. In the presented technique the EEG data are segmented into 4-second epochs. A k-nearest neighbor algorithm is employed to classify the EEG epochs into seizure and nonseizure groups. The performance of the presented technique is evaluated using the EEG data of five neonates. The results indicate that the proposed technique is superior to the other existing methods with 92.4 % good detection rate and 4.9 % false detection rate.

Original languageEnglish
Pages (from-to)137-146
Number of pages10
JournalInternational Journal of Engineering, Transactions A: Basics
Volume20
Issue number2
Publication statusPublished - 2007

Fingerprint

Electroencephalography

Keywords

  • Classification
  • EEG
  • Newborn
  • Seizure
  • Spike
  • Time-frequency

ASJC Scopus subject areas

  • Engineering(all)

Cite this

@article{f7b664c82d3449728f17cb0dee2cc657,
title = "Newborn EEG seizure detection based on interspike space distribution in the time-frequency domain",
abstract = "This paper presents a new time-frequency based EEG seizure detection method. This method uses the distribution of interspike intervals as a criterion for discriminating between seizure and nonseizure activities. To detect spikes in the EEG, the signal is mapped into the time-frequency domain. The high instantaneous energy of spikes is reflected as a localized energy in time-frequency domain. Histogram of successive spikes intervals is then used as a feature for seizure detection. In the presented technique the EEG data are segmented into 4-second epochs. A k-nearest neighbor algorithm is employed to classify the EEG epochs into seizure and nonseizure groups. The performance of the presented technique is evaluated using the EEG data of five neonates. The results indicate that the proposed technique is superior to the other existing methods with 92.4 {\%} good detection rate and 4.9 {\%} false detection rate.",
keywords = "Classification, EEG, Newborn, Seizure, Spike, Time-frequency",
author = "H. Hassanpour and M. Mesbah",
year = "2007",
language = "English",
volume = "20",
pages = "137--146",
journal = "International Journal of Engineering, Transactions B: Applications",
issn = "1728-144X",
publisher = "Materials and Energy Research Center",
number = "2",

}

TY - JOUR

T1 - Newborn EEG seizure detection based on interspike space distribution in the time-frequency domain

AU - Hassanpour, H.

AU - Mesbah, M.

PY - 2007

Y1 - 2007

N2 - This paper presents a new time-frequency based EEG seizure detection method. This method uses the distribution of interspike intervals as a criterion for discriminating between seizure and nonseizure activities. To detect spikes in the EEG, the signal is mapped into the time-frequency domain. The high instantaneous energy of spikes is reflected as a localized energy in time-frequency domain. Histogram of successive spikes intervals is then used as a feature for seizure detection. In the presented technique the EEG data are segmented into 4-second epochs. A k-nearest neighbor algorithm is employed to classify the EEG epochs into seizure and nonseizure groups. The performance of the presented technique is evaluated using the EEG data of five neonates. The results indicate that the proposed technique is superior to the other existing methods with 92.4 % good detection rate and 4.9 % false detection rate.

AB - This paper presents a new time-frequency based EEG seizure detection method. This method uses the distribution of interspike intervals as a criterion for discriminating between seizure and nonseizure activities. To detect spikes in the EEG, the signal is mapped into the time-frequency domain. The high instantaneous energy of spikes is reflected as a localized energy in time-frequency domain. Histogram of successive spikes intervals is then used as a feature for seizure detection. In the presented technique the EEG data are segmented into 4-second epochs. A k-nearest neighbor algorithm is employed to classify the EEG epochs into seizure and nonseizure groups. The performance of the presented technique is evaluated using the EEG data of five neonates. The results indicate that the proposed technique is superior to the other existing methods with 92.4 % good detection rate and 4.9 % false detection rate.

KW - Classification

KW - EEG

KW - Newborn

KW - Seizure

KW - Spike

KW - Time-frequency

UR - http://www.scopus.com/inward/record.url?scp=67649921333&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67649921333&partnerID=8YFLogxK

M3 - Article

VL - 20

SP - 137

EP - 146

JO - International Journal of Engineering, Transactions B: Applications

JF - International Journal of Engineering, Transactions B: Applications

SN - 1728-144X

IS - 2

ER -