A Study of Semi Supervised based approaches for Motor Imagery Signal Generation

Brain-computer interface provides an alternative way to communicate between the human brain and the external devices. Deep learning approaches have been widely used in various fields for feature extraction and classification task. However, the deep learning method requires a lot of data for training purpose. Due to the hectic calibration process, it is very difficult to collect large amount of EEG data. In such situations, deep neural network has proven very challenging in practice. This paper provides a comprehensive review of the various semi supervised approaches that have been used till now for the augmentation of motor imagery EEG data. Further, this research work has discussed about various research challenges faced by this field.