Accuracy Improvement of fNIRS based Motor Imagery Movement Classification by Standardized Common Spatial Pattern

Motor imagery event classification from functional near-infrared spectroscopy (fNIRS) is one of the most interesting problems of the current brain-computer interfaces (BCIs) challenges. Feature extraction from multiple channel fNIRS signal is always challenging due to its high dimensionality. To reduce the feature dimension, common spatial pattern (CSP) is an effective method. The present research work proposes an algorithm named by standardized common spatial pattern (SCSP) based filtering method for fNIRS based motor imagery classification. This work used both the conventional CSP method and the proposed SCSP method to extract the features from the fNIRS signal through spatial filtering. The results revealed that the proposed SCSP algorithm outperformed the conventional CSP method and channel-wise methods for classifying the two motor imagery event classifications. In addition, it can improve the classification accuracy up to 17% (in average) and 7%(in average) with respect to the channel-wise feature extraction method and conventional CSP algorithm, respectively.