Annals of Biomedical Engineering

, Volume 33, Issue 8, pp 1053–1070

Recognition of Motor Imagery Electroencephalography Using Independent Component Analysis and Machine Classifiers

Authors

  • Chih-I Hung
    • Institute of Radiological SciencesNational Yang-Ming University
    • Laboratory of Integrated Brain Research, Department of Medical Research and EducationTaipei Veterans General Hospital
  • Po-Lei Lee
    • Laboratory of Integrated Brain Research, Department of Medical Research and EducationTaipei Veterans General Hospital
    • Institute of Radiological SciencesNational Yang-Ming University
    • Laboratory of Integrated Brain Research, Department of Medical Research and EducationTaipei Veterans General Hospital
    • Institute of Health Informatics and Decision MakingSchool of Medicine, National Yang-Ming University
  • Li-Fen Chen
    • Laboratory of Integrated Brain Research, Department of Medical Research and EducationTaipei Veterans General Hospital
    • Center for NeuroscienceNational Yang-Ming University
  • Tzu-Chen Yeh
    • Laboratory of Integrated Brain Research, Department of Medical Research and EducationTaipei Veterans General Hospital
    • Faculty of MedicineSchool of Medicine, National Yang-Ming University
  • Jen-Chuen Hsieh
    • Laboratory of Integrated Brain Research, Department of Medical Research and EducationTaipei Veterans General Hospital
    • Institute of Health Informatics and Decision MakingSchool of Medicine, National Yang-Ming University
    • Faculty of MedicineSchool of Medicine, National Yang-Ming University
    • Institute of NeuroscienceSchool of Life Science, National Yang-Ming University
Article

DOI: 10.1007/s10439-005-5772-1

Cite this article as:
Hung, C., Lee, P., Wu, Y. et al. Ann Biomed Eng (2005) 33: 1053. doi:10.1007/s10439-005-5772-1

Abstract

Motor imagery electroencephalography (EEG), which embodies cortical potentials during mental simulation of left or right finger lifting tasks, can be used to provide neural input signals to activate a brain computer interface (BCI). The effectiveness of such an EEG-based BCI system relies on two indispensable components: distinguishable patterns of brain signals and accurate classifiers. This work aims to extract two reliable neural features, termed contralateral and ipsilateral rebound maps, by removing artifacts from motor imagery EEG based on independent component analysis (ICA), and to employ four classifiers to investigate the efficacy of rebound maps. Results demonstrate that, with the use of ICA, recognition rates for four classifiers (fisher linear discriminant (FLD), back-propagation neural network (BP-NN), radial-basis function neural network (RBF-NN), and support vector machine (SVM)) improved significantly, from 54%, 54%, 57% and 55% to 70.5%, 75.5%, 76.5% and 77.3%, respectively. In addition, the areas under the receiver operating characteristics (ROC) curve, which assess the quality of classification over a wide range of misclassification costs, also improved from .65, .60, .62, and .64 to .74, .76, .80 and .81, respectively.

Keywords

Brain computer interface (BCI)Rebound mapsFisher linear discriminant (FLD)Back-propagation neural network (BP-NN)Radial-basis function neural network (RBF-NN)Support vector machine (SVM)

Copyright information

© Biomedical Engineering Society 2005