Brain Computer Interface using Genetic Algorithm with modified Genome and Phenotype Structures

The human machine interface research in the light of modern fast computers and advanced sensors is taking new heights. The classification and processing of neural activity in the brain accessed by Magnetic Resonance Imaging (MRI), Positron Emission Tomography (PET), functional Magnetic Resonance Imaging (fMRI), Electrocorticography (ECoG), EEG Electroencephalogram (EEG) etc., are peeling off new paradigms for pattern recognition in human brain-machine interaction applications. In the present paper, an effective novel scheme based upon a synergetic approach employing the Genetic Algorithm (GA), Support Vector Machine and Wavelet packet transform for motor imagery classification and optimal Channel selection is proposed. GA with SVM acting as the objective function is employed for simultaneous selection of features and channels optimally. The binary population of GA is uniquely represented in three-dimensional structure and a new cross-over operator for GA are introduced. The new modified cross-over operator is proposed for the modified three-dimensional population. The ‘data set I’ of ‘BCI Competition IV’ is taken for evaluation of the efficacy of the proposed scheme. For subject ‘a’ accuracy is ۸۸.۹ ۶.۹ with ۱۰ channels, for subject ‘b’ accuracy is ۷۹.۲۰±۵.۳۶with ۱۱ channels, for subject ‘f’ accuracy is ۹۰.۵۰±۳.۵۶ with ۱۳ channels, and for subject ‘g’ accuracy is ۹۲.۲۳±۳.۲۱with ۱۲ channels. The proposed scheme outperforms in terms of classification accuracy for subjects ‘a, b, f, g’ and in terms of number of channels for subject ‘a’ and that for subject ‘b’ is same as reported earlier in literature. Therefore, proposed scheme contributes a significant development in terms of new three-dimensional representation of binary population for GA as well as significant new modification to the GA operators. The efficacy of the scheme is evident from the results presented in the paper for dataset under consideration.

Keywords:

Motor Imagery (M.I.) , Genetic Algorithm (GA) , Three Dimensional Population , Support Vector Machine (SVM)