EEG analysis of seizure patterns using visibility graphs for detection of generalized seizures

Research output: Contribution to journalArticleAcademicpeer-review

13 Citations (Scopus)
4 Downloads (Pure)

Abstract

Background The traditional EEG features in the time and frequency domain show limited seizure detection performance in the epileptic population with intellectual disability (ID). In addition, the influence of EEG seizure patterns on detection performance was less studied. New method A single-channel EEG signal can be mapped into visibility graphs (VGS), including basic visibility graph (VG), horizontal VG (HVG), and difference VG (DVG). These graphs were used to characterize different EEG seizure patterns. To demonstrate its effectiveness in identifying EEG seizure patterns and detecting generalized seizures, EEG recordings of 615 h on one EEG channel from 29 epileptic patients with ID were analyzed. Results A novel feature set with discriminative power for seizure detection was obtained by using the VGS method. The degree distributions (DDs) of DVG can clearly distinguish EEG of each seizure pattern. The degree entropy and power-law degree power in DVG were proposed here for the first time, and they show significant difference between seizure and non-seizure EEG. The connecting structure measured by HVG can better distinguish seizure EEG from background than those by VG and DVG. Comparison with existing method A traditional EEG feature set based on frequency analysis was used here as a benchmark feature set. With a support vector machine (SVM) classifier, the seizure detection performance of the benchmark feature set (sensitivity of 24%, FDt/h of 1.8s) can be improved by combining our proposed VGS features extracted from one EEG channel (sensitivity of 38%, FDt/h of 1.4s). Conclusions The proposed VGS-based features can help improve seizure detection for ID patients.

Original languageEnglish
Pages (from-to)85-94
Number of pages10
JournalJournal of Neuroscience Methods
Volume290
DOIs
Publication statusPublished - 1 Oct 2017

Fingerprint

Electroencephalography
Seizures
Intellectual Disability
Benchmarking
Entropy

Keywords

  • Difference VG (DVG)
  • EEG seizure pattern
  • Horizontal VG (HVG)
  • Intellectual disability
  • Seizure detection
  • Visibility graph (VG)

Cite this

@article{1273cf806127441884e5d3d575c3a6c6,
title = "EEG analysis of seizure patterns using visibility graphs for detection of generalized seizures",
abstract = "Background The traditional EEG features in the time and frequency domain show limited seizure detection performance in the epileptic population with intellectual disability (ID). In addition, the influence of EEG seizure patterns on detection performance was less studied. New method A single-channel EEG signal can be mapped into visibility graphs (VGS), including basic visibility graph (VG), horizontal VG (HVG), and difference VG (DVG). These graphs were used to characterize different EEG seizure patterns. To demonstrate its effectiveness in identifying EEG seizure patterns and detecting generalized seizures, EEG recordings of 615 h on one EEG channel from 29 epileptic patients with ID were analyzed. Results A novel feature set with discriminative power for seizure detection was obtained by using the VGS method. The degree distributions (DDs) of DVG can clearly distinguish EEG of each seizure pattern. The degree entropy and power-law degree power in DVG were proposed here for the first time, and they show significant difference between seizure and non-seizure EEG. The connecting structure measured by HVG can better distinguish seizure EEG from background than those by VG and DVG. Comparison with existing method A traditional EEG feature set based on frequency analysis was used here as a benchmark feature set. With a support vector machine (SVM) classifier, the seizure detection performance of the benchmark feature set (sensitivity of 24{\%}, FDt/h of 1.8s) can be improved by combining our proposed VGS features extracted from one EEG channel (sensitivity of 38{\%}, FDt/h of 1.4s). Conclusions The proposed VGS-based features can help improve seizure detection for ID patients.",
keywords = "Difference VG (DVG), EEG seizure pattern, Horizontal VG (HVG), Intellectual disability, Seizure detection, Visibility graph (VG)",
author = "Lei Wang and Xi Long and J.B.A.M. Arends and R.M. Aarts",
year = "2017",
month = "10",
day = "1",
doi = "10.1016/j.jneumeth.2017.07.013",
language = "English",
volume = "290",
pages = "85--94",
journal = "Journal of Neuroscience Methods",
issn = "0165-0270",
publisher = "Elsevier",

}

EEG analysis of seizure patterns using visibility graphs for detection of generalized seizures. / Wang, Lei; Long, Xi; Arends, J.B.A.M.; Aarts, R.M.

In: Journal of Neuroscience Methods, Vol. 290, 01.10.2017, p. 85-94.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - EEG analysis of seizure patterns using visibility graphs for detection of generalized seizures

AU - Wang, Lei

AU - Long, Xi

AU - Arends, J.B.A.M.

AU - Aarts, R.M.

PY - 2017/10/1

Y1 - 2017/10/1

N2 - Background The traditional EEG features in the time and frequency domain show limited seizure detection performance in the epileptic population with intellectual disability (ID). In addition, the influence of EEG seizure patterns on detection performance was less studied. New method A single-channel EEG signal can be mapped into visibility graphs (VGS), including basic visibility graph (VG), horizontal VG (HVG), and difference VG (DVG). These graphs were used to characterize different EEG seizure patterns. To demonstrate its effectiveness in identifying EEG seizure patterns and detecting generalized seizures, EEG recordings of 615 h on one EEG channel from 29 epileptic patients with ID were analyzed. Results A novel feature set with discriminative power for seizure detection was obtained by using the VGS method. The degree distributions (DDs) of DVG can clearly distinguish EEG of each seizure pattern. The degree entropy and power-law degree power in DVG were proposed here for the first time, and they show significant difference between seizure and non-seizure EEG. The connecting structure measured by HVG can better distinguish seizure EEG from background than those by VG and DVG. Comparison with existing method A traditional EEG feature set based on frequency analysis was used here as a benchmark feature set. With a support vector machine (SVM) classifier, the seizure detection performance of the benchmark feature set (sensitivity of 24%, FDt/h of 1.8s) can be improved by combining our proposed VGS features extracted from one EEG channel (sensitivity of 38%, FDt/h of 1.4s). Conclusions The proposed VGS-based features can help improve seizure detection for ID patients.

AB - Background The traditional EEG features in the time and frequency domain show limited seizure detection performance in the epileptic population with intellectual disability (ID). In addition, the influence of EEG seizure patterns on detection performance was less studied. New method A single-channel EEG signal can be mapped into visibility graphs (VGS), including basic visibility graph (VG), horizontal VG (HVG), and difference VG (DVG). These graphs were used to characterize different EEG seizure patterns. To demonstrate its effectiveness in identifying EEG seizure patterns and detecting generalized seizures, EEG recordings of 615 h on one EEG channel from 29 epileptic patients with ID were analyzed. Results A novel feature set with discriminative power for seizure detection was obtained by using the VGS method. The degree distributions (DDs) of DVG can clearly distinguish EEG of each seizure pattern. The degree entropy and power-law degree power in DVG were proposed here for the first time, and they show significant difference between seizure and non-seizure EEG. The connecting structure measured by HVG can better distinguish seizure EEG from background than those by VG and DVG. Comparison with existing method A traditional EEG feature set based on frequency analysis was used here as a benchmark feature set. With a support vector machine (SVM) classifier, the seizure detection performance of the benchmark feature set (sensitivity of 24%, FDt/h of 1.8s) can be improved by combining our proposed VGS features extracted from one EEG channel (sensitivity of 38%, FDt/h of 1.4s). Conclusions The proposed VGS-based features can help improve seizure detection for ID patients.

KW - Difference VG (DVG)

KW - EEG seizure pattern

KW - Horizontal VG (HVG)

KW - Intellectual disability

KW - Seizure detection

KW - Visibility graph (VG)

UR - http://www.scopus.com/inward/record.url?scp=85026682590&partnerID=8YFLogxK

U2 - 10.1016/j.jneumeth.2017.07.013

DO - 10.1016/j.jneumeth.2017.07.013

M3 - Article

C2 - 28734799

AN - SCOPUS:85026682590

VL - 290

SP - 85

EP - 94

JO - Journal of Neuroscience Methods

JF - Journal of Neuroscience Methods

SN - 0165-0270

ER -