EEG rhythm/channel selection for fuzzy rule-based alertness state characterization

Ahmed Al-Ani, Mostefa Mesbah

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

The aim of the paper is to automatically select the optimal EEG rhythm/channel combinations capable of classifying human alertness states. Four alertness states were considered, namely ‘engaged’, ‘calm’, ‘drowsy’ and ‘asleep’. The features used in the automatic selection are the energies associated with the conventional rhythms, (Formula presented.) and (Formula presented.), extracted from overlapping windows of the different EEG channels. The selection process consists of two stages. In the first stage, the optimal brain regions, represented by sets of EEG channels, are selected using a simple search technique based on support vector machine (SVM), extreme learning machine (ELM) and LDA classifiers. In the second stage, a fuzzy rule-based alertness classification system (FRBACS) is used to identify, from the previously selected EEG channels, the optimal features and their supports. The IF–THEN rules used in FRBACS are constructed using a novel differential evolution-based search algorithm particularly designed for this task. Each alertness state is represented by a set of IF–THEN rules whose antecedent parts contain EEG rhythm/channel combination. The selected spatio-frequency features were found to be good indicators of the different alertness states, as judged by the classification performance of the FRBACS that was found to be comparable to those of the SVM, ELM and LDA classifiers. Moreover, the proposed classification system has the advantage of revealing simple and easy to interpret decision rules associated with each of the alertness states.

Original languageEnglish
Pages (from-to)1-11
Number of pages11
JournalNeural Computing and Applications
DOIs
Publication statusAccepted/In press - Dec 30 2016

Fingerprint

Fuzzy rules
Electroencephalography
Support vector machines
Learning systems
Classifiers
Brain

Keywords

  • Alertness classification
  • Differential evolution
  • Drowsiness
  • EEG
  • Feature selection
  • Fuzzy rule-based classification system

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence

Cite this

EEG rhythm/channel selection for fuzzy rule-based alertness state characterization. / Al-Ani, Ahmed; Mesbah, Mostefa.

In: Neural Computing and Applications, 30.12.2016, p. 1-11.

Research output: Contribution to journalArticle

@article{7d0b7540b347469b98c976ffd30e2c7c,
title = "EEG rhythm/channel selection for fuzzy rule-based alertness state characterization",
abstract = "The aim of the paper is to automatically select the optimal EEG rhythm/channel combinations capable of classifying human alertness states. Four alertness states were considered, namely ‘engaged’, ‘calm’, ‘drowsy’ and ‘asleep’. The features used in the automatic selection are the energies associated with the conventional rhythms, (Formula presented.) and (Formula presented.), extracted from overlapping windows of the different EEG channels. The selection process consists of two stages. In the first stage, the optimal brain regions, represented by sets of EEG channels, are selected using a simple search technique based on support vector machine (SVM), extreme learning machine (ELM) and LDA classifiers. In the second stage, a fuzzy rule-based alertness classification system (FRBACS) is used to identify, from the previously selected EEG channels, the optimal features and their supports. The IF–THEN rules used in FRBACS are constructed using a novel differential evolution-based search algorithm particularly designed for this task. Each alertness state is represented by a set of IF–THEN rules whose antecedent parts contain EEG rhythm/channel combination. The selected spatio-frequency features were found to be good indicators of the different alertness states, as judged by the classification performance of the FRBACS that was found to be comparable to those of the SVM, ELM and LDA classifiers. Moreover, the proposed classification system has the advantage of revealing simple and easy to interpret decision rules associated with each of the alertness states.",
keywords = "Alertness classification, Differential evolution, Drowsiness, EEG, Feature selection, Fuzzy rule-based classification system",
author = "Ahmed Al-Ani and Mostefa Mesbah",
year = "2016",
month = "12",
day = "30",
doi = "10.1007/s00521-016-2835-1",
language = "English",
pages = "1--11",
journal = "Neural Computing and Applications",
issn = "0941-0643",
publisher = "Springer London",

}

TY - JOUR

T1 - EEG rhythm/channel selection for fuzzy rule-based alertness state characterization

AU - Al-Ani, Ahmed

AU - Mesbah, Mostefa

PY - 2016/12/30

Y1 - 2016/12/30

N2 - The aim of the paper is to automatically select the optimal EEG rhythm/channel combinations capable of classifying human alertness states. Four alertness states were considered, namely ‘engaged’, ‘calm’, ‘drowsy’ and ‘asleep’. The features used in the automatic selection are the energies associated with the conventional rhythms, (Formula presented.) and (Formula presented.), extracted from overlapping windows of the different EEG channels. The selection process consists of two stages. In the first stage, the optimal brain regions, represented by sets of EEG channels, are selected using a simple search technique based on support vector machine (SVM), extreme learning machine (ELM) and LDA classifiers. In the second stage, a fuzzy rule-based alertness classification system (FRBACS) is used to identify, from the previously selected EEG channels, the optimal features and their supports. The IF–THEN rules used in FRBACS are constructed using a novel differential evolution-based search algorithm particularly designed for this task. Each alertness state is represented by a set of IF–THEN rules whose antecedent parts contain EEG rhythm/channel combination. The selected spatio-frequency features were found to be good indicators of the different alertness states, as judged by the classification performance of the FRBACS that was found to be comparable to those of the SVM, ELM and LDA classifiers. Moreover, the proposed classification system has the advantage of revealing simple and easy to interpret decision rules associated with each of the alertness states.

AB - The aim of the paper is to automatically select the optimal EEG rhythm/channel combinations capable of classifying human alertness states. Four alertness states were considered, namely ‘engaged’, ‘calm’, ‘drowsy’ and ‘asleep’. The features used in the automatic selection are the energies associated with the conventional rhythms, (Formula presented.) and (Formula presented.), extracted from overlapping windows of the different EEG channels. The selection process consists of two stages. In the first stage, the optimal brain regions, represented by sets of EEG channels, are selected using a simple search technique based on support vector machine (SVM), extreme learning machine (ELM) and LDA classifiers. In the second stage, a fuzzy rule-based alertness classification system (FRBACS) is used to identify, from the previously selected EEG channels, the optimal features and their supports. The IF–THEN rules used in FRBACS are constructed using a novel differential evolution-based search algorithm particularly designed for this task. Each alertness state is represented by a set of IF–THEN rules whose antecedent parts contain EEG rhythm/channel combination. The selected spatio-frequency features were found to be good indicators of the different alertness states, as judged by the classification performance of the FRBACS that was found to be comparable to those of the SVM, ELM and LDA classifiers. Moreover, the proposed classification system has the advantage of revealing simple and easy to interpret decision rules associated with each of the alertness states.

KW - Alertness classification

KW - Differential evolution

KW - Drowsiness

KW - EEG

KW - Feature selection

KW - Fuzzy rule-based classification system

UR - http://www.scopus.com/inward/record.url?scp=85007492413&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85007492413&partnerID=8YFLogxK

U2 - 10.1007/s00521-016-2835-1

DO - 10.1007/s00521-016-2835-1

M3 - Article

AN - SCOPUS:85007492413

SP - 1

EP - 11

JO - Neural Computing and Applications

JF - Neural Computing and Applications

SN - 0941-0643

ER -