Classification of Four-Class Motor Imagery EEG Data Using Spatial Filtering

An efficient discrimination method of different brain states is important in order to implement a suitable system. This paper applies the One Versus the Rest CSP to the EEG data recorded during motor imagery tasks consisting of four different classes, namely the imagination of right hand, left hand, foot and tongue movements. The best correct classification results for the three subjects are 73.3%, 81.7%, and 76.7%.

[1]  G Pfurtscheller,et al.  Mapping of event-related desynchronization and type of derivation. , 1988, Electroencephalography and clinical neurophysiology.

[2]  G. Pfurtscheller,et al.  Motor imagery activates primary sensorimotor area in humans , 1997, Neuroscience Letters.

[3]  J R Wolpaw,et al.  Spatial filter selection for EEG-based communication. , 1997, Electroencephalography and clinical neurophysiology.

[4]  Klaus-Robert Müller,et al.  Boosting bit rates in noninvasive EEG single-trial classifications by feature combination and multiclass paradigms , 2004, IEEE Transactions on Biomedical Engineering.