Classification of brain signal (EEG) induced by shape-analogous letter perception

Rohit Bose, Sim Kuan Goh, Kian F. Wong, Nitish Thakor, Anastasios Bezerianos, Junhua Li

Research output: Contribution to journalArticle

Abstract

Visual perception of English letters involves different underlying brain processes including brain activity alteration in multiple frequency bands. However, shape analogous letters elicit brain activities which are not obviously distinct and it is therefore difficult to differentiate those activities. In order to address discriminative feasibility and classification performance of the perception of shape-analogous letters, we performed an experiment in where EEG signals were obtained from 20 subjects while they were perceiving shape analogous letters (i.e., ‘p’, ‘q’, ‘b’, and ‘d’). Spectral power densities from five typical frequency bands (i.e., delta, theta, alpha, beta and gamma) were extracted as features, which were then classified by either individual widely-used classifiers, namely k-Nearest Neighbors (kNN), Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), Random Forest (RF) and AdaBoost (ADA), or an ensemble of some of them. The F-score was employed to select most discriminative features so that the dimension of features was reduced. The results showed that the RF achieved the highest accuracy of 74.1% in the case of multi-class classification. In the case of binary classification, the best performance (Accuracy 86.39%) was achieved by the RF classifier in terms of average accuracy across all possible pairs of the letters. In addition, we employed decision fusion strategy to exert complementary strengths of different classifiers. The results demonstrated that the performance was elevated from 74.10% to 76.63% for the multi-class classification and from 86.39% to 88.08% for the binary class classification.

Original languageEnglish (US)
Article number100992
JournalAdvanced Engineering Informatics
Volume42
DOIs
StatePublished - Oct 2019
Externally publishedYes

Fingerprint

Electroencephalography
Brain
Classifiers
Frequency bands
Adaptive boosting
Power spectral density
Discriminant analysis
Support vector machines
Fusion reactions
Experiments

Keywords

  • AdaBoost
  • Decision level fusion
  • Electroencephalography (EEG)
  • F-score
  • k-Nearest Neighbors
  • Linear Discriminant Analysis
  • Multi-class classification
  • Random Forest
  • Shape analogous letters
  • Support Vector Machine

ASJC Scopus subject areas

  • Information Systems
  • Artificial Intelligence

Cite this

Classification of brain signal (EEG) induced by shape-analogous letter perception. / Bose, Rohit; Goh, Sim Kuan; Wong, Kian F.; Thakor, Nitish; Bezerianos, Anastasios; Li, Junhua.

In: Advanced Engineering Informatics, Vol. 42, 100992, 10.2019.

Research output: Contribution to journalArticle

Bose, Rohit ; Goh, Sim Kuan ; Wong, Kian F. ; Thakor, Nitish ; Bezerianos, Anastasios ; Li, Junhua. / Classification of brain signal (EEG) induced by shape-analogous letter perception. In: Advanced Engineering Informatics. 2019 ; Vol. 42.
@article{92a1b16da4d645499fe620a31e69c4dc,
title = "Classification of brain signal (EEG) induced by shape-analogous letter perception",
abstract = "Visual perception of English letters involves different underlying brain processes including brain activity alteration in multiple frequency bands. However, shape analogous letters elicit brain activities which are not obviously distinct and it is therefore difficult to differentiate those activities. In order to address discriminative feasibility and classification performance of the perception of shape-analogous letters, we performed an experiment in where EEG signals were obtained from 20 subjects while they were perceiving shape analogous letters (i.e., ‘p’, ‘q’, ‘b’, and ‘d’). Spectral power densities from five typical frequency bands (i.e., delta, theta, alpha, beta and gamma) were extracted as features, which were then classified by either individual widely-used classifiers, namely k-Nearest Neighbors (kNN), Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), Random Forest (RF) and AdaBoost (ADA), or an ensemble of some of them. The F-score was employed to select most discriminative features so that the dimension of features was reduced. The results showed that the RF achieved the highest accuracy of 74.1{\%} in the case of multi-class classification. In the case of binary classification, the best performance (Accuracy 86.39{\%}) was achieved by the RF classifier in terms of average accuracy across all possible pairs of the letters. In addition, we employed decision fusion strategy to exert complementary strengths of different classifiers. The results demonstrated that the performance was elevated from 74.10{\%} to 76.63{\%} for the multi-class classification and from 86.39{\%} to 88.08{\%} for the binary class classification.",
keywords = "AdaBoost, Decision level fusion, Electroencephalography (EEG), F-score, k-Nearest Neighbors, Linear Discriminant Analysis, Multi-class classification, Random Forest, Shape analogous letters, Support Vector Machine",
author = "Rohit Bose and Goh, {Sim Kuan} and Wong, {Kian F.} and Nitish Thakor and Anastasios Bezerianos and Junhua Li",
year = "2019",
month = "10",
doi = "10.1016/j.aei.2019.100992",
language = "English (US)",
volume = "42",
journal = "Advanced Engineering Informatics",
issn = "1474-0346",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Classification of brain signal (EEG) induced by shape-analogous letter perception

AU - Bose, Rohit

AU - Goh, Sim Kuan

AU - Wong, Kian F.

AU - Thakor, Nitish

AU - Bezerianos, Anastasios

AU - Li, Junhua

PY - 2019/10

Y1 - 2019/10

N2 - Visual perception of English letters involves different underlying brain processes including brain activity alteration in multiple frequency bands. However, shape analogous letters elicit brain activities which are not obviously distinct and it is therefore difficult to differentiate those activities. In order to address discriminative feasibility and classification performance of the perception of shape-analogous letters, we performed an experiment in where EEG signals were obtained from 20 subjects while they were perceiving shape analogous letters (i.e., ‘p’, ‘q’, ‘b’, and ‘d’). Spectral power densities from five typical frequency bands (i.e., delta, theta, alpha, beta and gamma) were extracted as features, which were then classified by either individual widely-used classifiers, namely k-Nearest Neighbors (kNN), Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), Random Forest (RF) and AdaBoost (ADA), or an ensemble of some of them. The F-score was employed to select most discriminative features so that the dimension of features was reduced. The results showed that the RF achieved the highest accuracy of 74.1% in the case of multi-class classification. In the case of binary classification, the best performance (Accuracy 86.39%) was achieved by the RF classifier in terms of average accuracy across all possible pairs of the letters. In addition, we employed decision fusion strategy to exert complementary strengths of different classifiers. The results demonstrated that the performance was elevated from 74.10% to 76.63% for the multi-class classification and from 86.39% to 88.08% for the binary class classification.

AB - Visual perception of English letters involves different underlying brain processes including brain activity alteration in multiple frequency bands. However, shape analogous letters elicit brain activities which are not obviously distinct and it is therefore difficult to differentiate those activities. In order to address discriminative feasibility and classification performance of the perception of shape-analogous letters, we performed an experiment in where EEG signals were obtained from 20 subjects while they were perceiving shape analogous letters (i.e., ‘p’, ‘q’, ‘b’, and ‘d’). Spectral power densities from five typical frequency bands (i.e., delta, theta, alpha, beta and gamma) were extracted as features, which were then classified by either individual widely-used classifiers, namely k-Nearest Neighbors (kNN), Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), Random Forest (RF) and AdaBoost (ADA), or an ensemble of some of them. The F-score was employed to select most discriminative features so that the dimension of features was reduced. The results showed that the RF achieved the highest accuracy of 74.1% in the case of multi-class classification. In the case of binary classification, the best performance (Accuracy 86.39%) was achieved by the RF classifier in terms of average accuracy across all possible pairs of the letters. In addition, we employed decision fusion strategy to exert complementary strengths of different classifiers. The results demonstrated that the performance was elevated from 74.10% to 76.63% for the multi-class classification and from 86.39% to 88.08% for the binary class classification.

KW - AdaBoost

KW - Decision level fusion

KW - Electroencephalography (EEG)

KW - F-score

KW - k-Nearest Neighbors

KW - Linear Discriminant Analysis

KW - Multi-class classification

KW - Random Forest

KW - Shape analogous letters

KW - Support Vector Machine

UR - http://www.scopus.com/inward/record.url?scp=85073204017&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073204017&partnerID=8YFLogxK

U2 - 10.1016/j.aei.2019.100992

DO - 10.1016/j.aei.2019.100992

M3 - Article

AN - SCOPUS:85073204017

VL - 42

JO - Advanced Engineering Informatics

JF - Advanced Engineering Informatics

SN - 1474-0346

M1 - 100992

ER -