EEG Classification of Imaginary Lower Limb Stepping Movements Based on Fuzzy Support Vector Machine with Kernel-Induced Membership Function

Wei Chun Hsu, Li Fong Lin, Chun Wei Chou, Yu Tsung Hsiao, Yi Hung Liu

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

Although various kinds of motor imageries have been used for BCI applications, imaginary lower limb stepping movement has not been studied yet. The purpose of this study is to investigate the possibilities of using electroencephalography (EEG) signal to classify imaginary lower limb stepping movements and to design a robust motor imagery classifier based on support vector machine (SVM). A cue-based experimental paradigm is designed to record nine-channel EEG associated with imaginary left leg stepping (L-stepping) and right leg stepping (R-stepping) movements from eight healthy subjects. Features including band powers (BPs), common spatial pattern (CSP), and a filter-bank CSP (FB-CSP) were extracted from the recorded EEG. Fuzzy SVM (FSVM) is introduced to this study to classify L-stepping and R-stepping imageries. We propose a novel kernel-induced membership function to address the issue of data relative importance assignment. The FSVM with the membership function suggested in the original work of FSVM (Type-I FSVM) and the FSVM with the one we proposed (Type-II FSVM) is compared. Results indicated that the classification accuracies based on BP features are near the chance level (~50 %). Both alpha-band CSP (71.25 %) and FB-CSP (75.63 %) gave acceptable results as a simple k-NN classifier is performed. Results show that both types of FSVM performed better than the conventional SVM. Also, Type-II FSVM outperforms Type-I FSVM, especially when the alpha-CSP feature is employed, where the improvement in error reduction rate is over 15 %. The highest average L-stepping versus R-stepping classification accuracy over the eight subjects is achieved (86.25 % in single-trial analysis) by FB-CSP and FSVM-II. The high classification result suggests the feasibility of using lower limb stepping imagery to develop a BCI that can control devices or might be able to serve as a neurofeedback tool for users who need lower limb stepping imagery training for gait function improvement.

Original languageEnglish
Pages (from-to)566-579
Number of pages14
JournalInternational Journal of Fuzzy Systems
Volume19
Issue number2
DOIs
Publication statusPublished - Apr 1 2017

Fingerprint

Electroencephalography
Filter banks
Membership functions
Membership Function
Support vector machines
Support Vector Machine
kernel
Classifiers
Spatial Pattern
Filter Banks
Classify
Classifier
Movement
Error Reduction
Gait
Assignment
Paradigm
Imagery

Keywords

  • Brain–computer interface
  • Common spatial pattern
  • EEG
  • Lower limb stepping
  • Motor imagery
  • Support vector machine

ASJC Scopus subject areas

  • Software
  • Theoretical Computer Science
  • Computational Theory and Mathematics
  • Artificial Intelligence

Cite this

EEG Classification of Imaginary Lower Limb Stepping Movements Based on Fuzzy Support Vector Machine with Kernel-Induced Membership Function. / Hsu, Wei Chun; Lin, Li Fong; Chou, Chun Wei; Hsiao, Yu Tsung; Liu, Yi Hung.

In: International Journal of Fuzzy Systems, Vol. 19, No. 2, 01.04.2017, p. 566-579.

Research output: Contribution to journalArticle

@article{d634a54e90e24fca9d0bdbb249250ee4,
title = "EEG Classification of Imaginary Lower Limb Stepping Movements Based on Fuzzy Support Vector Machine with Kernel-Induced Membership Function",
abstract = "Although various kinds of motor imageries have been used for BCI applications, imaginary lower limb stepping movement has not been studied yet. The purpose of this study is to investigate the possibilities of using electroencephalography (EEG) signal to classify imaginary lower limb stepping movements and to design a robust motor imagery classifier based on support vector machine (SVM). A cue-based experimental paradigm is designed to record nine-channel EEG associated with imaginary left leg stepping (L-stepping) and right leg stepping (R-stepping) movements from eight healthy subjects. Features including band powers (BPs), common spatial pattern (CSP), and a filter-bank CSP (FB-CSP) were extracted from the recorded EEG. Fuzzy SVM (FSVM) is introduced to this study to classify L-stepping and R-stepping imageries. We propose a novel kernel-induced membership function to address the issue of data relative importance assignment. The FSVM with the membership function suggested in the original work of FSVM (Type-I FSVM) and the FSVM with the one we proposed (Type-II FSVM) is compared. Results indicated that the classification accuracies based on BP features are near the chance level (~50 {\%}). Both alpha-band CSP (71.25 {\%}) and FB-CSP (75.63 {\%}) gave acceptable results as a simple k-NN classifier is performed. Results show that both types of FSVM performed better than the conventional SVM. Also, Type-II FSVM outperforms Type-I FSVM, especially when the alpha-CSP feature is employed, where the improvement in error reduction rate is over 15 {\%}. The highest average L-stepping versus R-stepping classification accuracy over the eight subjects is achieved (86.25 {\%} in single-trial analysis) by FB-CSP and FSVM-II. The high classification result suggests the feasibility of using lower limb stepping imagery to develop a BCI that can control devices or might be able to serve as a neurofeedback tool for users who need lower limb stepping imagery training for gait function improvement.",
keywords = "Brain–computer interface, Common spatial pattern, EEG, Lower limb stepping, Motor imagery, Support vector machine",
author = "Hsu, {Wei Chun} and Lin, {Li Fong} and Chou, {Chun Wei} and Hsiao, {Yu Tsung} and Liu, {Yi Hung}",
year = "2017",
month = "4",
day = "1",
doi = "10.1007/s40815-016-0259-9",
language = "English",
volume = "19",
pages = "566--579",
journal = "International Journal of Fuzzy Systems",
issn = "1562-2479",
publisher = "Chinese Fuzzy Systems Association",
number = "2",

}

TY - JOUR

T1 - EEG Classification of Imaginary Lower Limb Stepping Movements Based on Fuzzy Support Vector Machine with Kernel-Induced Membership Function

AU - Hsu, Wei Chun

AU - Lin, Li Fong

AU - Chou, Chun Wei

AU - Hsiao, Yu Tsung

AU - Liu, Yi Hung

PY - 2017/4/1

Y1 - 2017/4/1

N2 - Although various kinds of motor imageries have been used for BCI applications, imaginary lower limb stepping movement has not been studied yet. The purpose of this study is to investigate the possibilities of using electroencephalography (EEG) signal to classify imaginary lower limb stepping movements and to design a robust motor imagery classifier based on support vector machine (SVM). A cue-based experimental paradigm is designed to record nine-channel EEG associated with imaginary left leg stepping (L-stepping) and right leg stepping (R-stepping) movements from eight healthy subjects. Features including band powers (BPs), common spatial pattern (CSP), and a filter-bank CSP (FB-CSP) were extracted from the recorded EEG. Fuzzy SVM (FSVM) is introduced to this study to classify L-stepping and R-stepping imageries. We propose a novel kernel-induced membership function to address the issue of data relative importance assignment. The FSVM with the membership function suggested in the original work of FSVM (Type-I FSVM) and the FSVM with the one we proposed (Type-II FSVM) is compared. Results indicated that the classification accuracies based on BP features are near the chance level (~50 %). Both alpha-band CSP (71.25 %) and FB-CSP (75.63 %) gave acceptable results as a simple k-NN classifier is performed. Results show that both types of FSVM performed better than the conventional SVM. Also, Type-II FSVM outperforms Type-I FSVM, especially when the alpha-CSP feature is employed, where the improvement in error reduction rate is over 15 %. The highest average L-stepping versus R-stepping classification accuracy over the eight subjects is achieved (86.25 % in single-trial analysis) by FB-CSP and FSVM-II. The high classification result suggests the feasibility of using lower limb stepping imagery to develop a BCI that can control devices or might be able to serve as a neurofeedback tool for users who need lower limb stepping imagery training for gait function improvement.

AB - Although various kinds of motor imageries have been used for BCI applications, imaginary lower limb stepping movement has not been studied yet. The purpose of this study is to investigate the possibilities of using electroencephalography (EEG) signal to classify imaginary lower limb stepping movements and to design a robust motor imagery classifier based on support vector machine (SVM). A cue-based experimental paradigm is designed to record nine-channel EEG associated with imaginary left leg stepping (L-stepping) and right leg stepping (R-stepping) movements from eight healthy subjects. Features including band powers (BPs), common spatial pattern (CSP), and a filter-bank CSP (FB-CSP) were extracted from the recorded EEG. Fuzzy SVM (FSVM) is introduced to this study to classify L-stepping and R-stepping imageries. We propose a novel kernel-induced membership function to address the issue of data relative importance assignment. The FSVM with the membership function suggested in the original work of FSVM (Type-I FSVM) and the FSVM with the one we proposed (Type-II FSVM) is compared. Results indicated that the classification accuracies based on BP features are near the chance level (~50 %). Both alpha-band CSP (71.25 %) and FB-CSP (75.63 %) gave acceptable results as a simple k-NN classifier is performed. Results show that both types of FSVM performed better than the conventional SVM. Also, Type-II FSVM outperforms Type-I FSVM, especially when the alpha-CSP feature is employed, where the improvement in error reduction rate is over 15 %. The highest average L-stepping versus R-stepping classification accuracy over the eight subjects is achieved (86.25 % in single-trial analysis) by FB-CSP and FSVM-II. The high classification result suggests the feasibility of using lower limb stepping imagery to develop a BCI that can control devices or might be able to serve as a neurofeedback tool for users who need lower limb stepping imagery training for gait function improvement.

KW - Brain–computer interface

KW - Common spatial pattern

KW - EEG

KW - Lower limb stepping

KW - Motor imagery

KW - Support vector machine

UR - http://www.scopus.com/inward/record.url?scp=85016927197&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85016927197&partnerID=8YFLogxK

U2 - 10.1007/s40815-016-0259-9

DO - 10.1007/s40815-016-0259-9

M3 - Article

AN - SCOPUS:85016927197

VL - 19

SP - 566

EP - 579

JO - International Journal of Fuzzy Systems

JF - International Journal of Fuzzy Systems

SN - 1562-2479

IS - 2

ER -