A Novel Feature Selection Method For Motor Imagery-Based Brain-Computer Interfaces

Feature selection in brain-computer interface (BCI) systems is an important stage that can improve the system performance especially in the presence of a big number of features extracted. In this paper, a new feature selection method is proposed which is a combination of PCA and mRMR. CSP and SVM are used for feature extraction and classification, respectively. The results show that our proposed method for feature selection has a better performance than PCA and mRMR methods.