Brain Data Mining Framework Involving Entropy Topography and Deep Learning.

Mining large scale brain signal data using artificial intelligence offers an unparalleled chance to investigate the dynamics of the brain in neurological disorders diagnosis. Electroencephalography (EEG) produces amulti-channel time-series large scale brain signal data recorded from scalp and visually analyzed by expert clinicians for abnormality detection. It is time-consuming, error-prone, subjective and has reliability issues. Thus, there is always a need of automated mining system for brain signal data to detect abnormality from those large volume data. This study presents an entropy topography with deep learning-based technique to solve the above mentioned issues. We have used shannon entropy to extract entropy values from EEG signal and plotted them to produce the topographic image. Then those images are trained and classified using our proposed convolutional neural network. We have tested it on two EEG datasets of schizophrenia disorder and the results showed that the proposed method can be used for brain signal data mining purposes.