Classification of left and right hand motor imagery EEG signals by using deep neural networks

The brain-computer interface (BCI) is one of the most promising technologies that allows us to establish a relationship between brain and devices. In this study, three-channel EEG signals collected from nine subjects performing two motor imagery tasks are classified using different deep neural network (DNN) based approaches called framework 1 (FW1) 2 (FW2). proposed frameworks were evaluated BCI Competition IV-IIb dataset. FW1, raw data directly presented without any pre-processing. FW2, first filtered with five band pass filters fifth order (Butterworth), then common spatial patterns (CSP) method, which introduces additional pseudo channels, applied signals. Two experiments conducted for each framework. experiment, unique DNN trained subject, in second experiment only combination training sets all subjects. performance compared terms average accuracy. According simulation results, we did not observe significant difference classification accuracies obtained experiments. Therefore, concluded that, by use DNNs do need train several subject-specific networks requires high computational loads. On other hand, observed significantly improves filtering extracting features CSP pre-processes.