This study proposes an electroencephalographic (EEG) analysis system for brain-computer interface applications. With the combination of neuro-fuzzy prediction, multiscale synchronization features are applied for feature extract ion in motor imagery (MI) analysis. The features are extracted from EEG signals recorded from subjects performing left and right MI. Time-series predictions are performed by training two adaptive neuro-fuzzy inference systems for respective left and right MI data. Features are then calculated from the difference of multiscale coherence and phase-locking-value features between the predicted and actual signals through a window of EEG signal s. Finally, a support vector machine classifier is used for classification. The performance of the proposed system is compared to that of two popular approaches on six subjects from two data sets. The results indicate that the proposed system is promising for MI classification.
