Automated seizure detection using limited-channel EEG and non-linear dimension reduction

Javad Birjandtalab, Maziyar Baran Pouyan, Diana Cogan, Mehrdad Nourani, Jay Harvey

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

Electroencephalography (EEG) is an essential component in evaluation of epilepsy. However, full-channel EEG signals recorded from 18 to 23 electrodes on the scalp is neither wearable nor computationally effective. This paper presents advantages of both channel selection and nonlinear dimension reduction for accurate automatic seizure detection. We first extract the frequency domain features from the full-channel EEG signals. Then, we use a random forest algorithm to determine which channels contribute the most in discriminating seizure from non-seizure events. Next, we apply a non-linear dimension reduction technique to capture the relationship among data elements and map them in low dimension. Finally, we apply a KNN classifier technique to discriminate between seizure and non-seizure events. The experimental results for 23 patients show that our proposed approach outperforms other techniques in terms of accuracy. It also visualizes long-term data in 2D to enhance physician cognition of occurrence and disease progression.

Original languageEnglish (US)
Pages (from-to)49-58
Number of pages10
JournalComputers in Biology and Medicine
Volume82
DOIs
StatePublished - Mar 1 2017

Fingerprint

Electroencephalography
Seizures
Scalp
Cognition
Disease Progression
Epilepsy
Electrodes
Classifiers
Physicians

Keywords

  • Channel selection
  • EEG signals
  • Feature extraction
  • Nonlinear dimension reduction
  • Random forest
  • Seizure detection.

ASJC Scopus subject areas

  • Computer Science Applications
  • Health Informatics

Cite this

Automated seizure detection using limited-channel EEG and non-linear dimension reduction. / Birjandtalab, Javad; Baran Pouyan, Maziyar; Cogan, Diana; Nourani, Mehrdad; Harvey, Jay.

In: Computers in Biology and Medicine, Vol. 82, 01.03.2017, p. 49-58.

Research output: Contribution to journalArticle

Birjandtalab, Javad ; Baran Pouyan, Maziyar ; Cogan, Diana ; Nourani, Mehrdad ; Harvey, Jay. / Automated seizure detection using limited-channel EEG and non-linear dimension reduction. In: Computers in Biology and Medicine. 2017 ; Vol. 82. pp. 49-58.
@article{fcf7d70390844f3f92907070ff4aeb49,
title = "Automated seizure detection using limited-channel EEG and non-linear dimension reduction",
abstract = "Electroencephalography (EEG) is an essential component in evaluation of epilepsy. However, full-channel EEG signals recorded from 18 to 23 electrodes on the scalp is neither wearable nor computationally effective. This paper presents advantages of both channel selection and nonlinear dimension reduction for accurate automatic seizure detection. We first extract the frequency domain features from the full-channel EEG signals. Then, we use a random forest algorithm to determine which channels contribute the most in discriminating seizure from non-seizure events. Next, we apply a non-linear dimension reduction technique to capture the relationship among data elements and map them in low dimension. Finally, we apply a KNN classifier technique to discriminate between seizure and non-seizure events. The experimental results for 23 patients show that our proposed approach outperforms other techniques in terms of accuracy. It also visualizes long-term data in 2D to enhance physician cognition of occurrence and disease progression.",
keywords = "Channel selection, EEG signals, Feature extraction, Nonlinear dimension reduction, Random forest, Seizure detection.",
author = "Javad Birjandtalab and {Baran Pouyan}, Maziyar and Diana Cogan and Mehrdad Nourani and Jay Harvey",
year = "2017",
month = "3",
day = "1",
doi = "10.1016/j.compbiomed.2017.01.011",
language = "English (US)",
volume = "82",
pages = "49--58",
journal = "Computers in Biology and Medicine",
issn = "0010-4825",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Automated seizure detection using limited-channel EEG and non-linear dimension reduction

AU - Birjandtalab, Javad

AU - Baran Pouyan, Maziyar

AU - Cogan, Diana

AU - Nourani, Mehrdad

AU - Harvey, Jay

PY - 2017/3/1

Y1 - 2017/3/1

N2 - Electroencephalography (EEG) is an essential component in evaluation of epilepsy. However, full-channel EEG signals recorded from 18 to 23 electrodes on the scalp is neither wearable nor computationally effective. This paper presents advantages of both channel selection and nonlinear dimension reduction for accurate automatic seizure detection. We first extract the frequency domain features from the full-channel EEG signals. Then, we use a random forest algorithm to determine which channels contribute the most in discriminating seizure from non-seizure events. Next, we apply a non-linear dimension reduction technique to capture the relationship among data elements and map them in low dimension. Finally, we apply a KNN classifier technique to discriminate between seizure and non-seizure events. The experimental results for 23 patients show that our proposed approach outperforms other techniques in terms of accuracy. It also visualizes long-term data in 2D to enhance physician cognition of occurrence and disease progression.

AB - Electroencephalography (EEG) is an essential component in evaluation of epilepsy. However, full-channel EEG signals recorded from 18 to 23 electrodes on the scalp is neither wearable nor computationally effective. This paper presents advantages of both channel selection and nonlinear dimension reduction for accurate automatic seizure detection. We first extract the frequency domain features from the full-channel EEG signals. Then, we use a random forest algorithm to determine which channels contribute the most in discriminating seizure from non-seizure events. Next, we apply a non-linear dimension reduction technique to capture the relationship among data elements and map them in low dimension. Finally, we apply a KNN classifier technique to discriminate between seizure and non-seizure events. The experimental results for 23 patients show that our proposed approach outperforms other techniques in terms of accuracy. It also visualizes long-term data in 2D to enhance physician cognition of occurrence and disease progression.

KW - Channel selection

KW - EEG signals

KW - Feature extraction

KW - Nonlinear dimension reduction

KW - Random forest

KW - Seizure detection.

UR - http://www.scopus.com/inward/record.url?scp=85011260926&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85011260926&partnerID=8YFLogxK

U2 - 10.1016/j.compbiomed.2017.01.011

DO - 10.1016/j.compbiomed.2017.01.011

M3 - Article

C2 - 28161592

AN - SCOPUS:85011260926

VL - 82

SP - 49

EP - 58

JO - Computers in Biology and Medicine

JF - Computers in Biology and Medicine

SN - 0010-4825

ER -