A review of the classification of neuroscience problems with the help of deep learning framework

Electroencephalographic signals (EEG signals) processing has become very popular nowadays due to its effectiveness in dealing with and treating various disorders associated with the field of neuroscience. As per the recent trends, deep learning has shown many promising results as compared to machine learning due to its ability to extract end-to-end features automatically from the raw input data and subsequently providing better performance in classification results. This paper has reviewed research papers that implemented various deep learning methodologies i.e., CNN, R-CNN, LSTM, GAN, etc. for the classification of EEG signals specific to epilepsy, sleep stage, and mental stages disorders. The review has been carried out considering various parameters i.e., objectives, datasets, models, results, etc. This paper has also discussed the common challenges associated with EEG signals classification related to neuroscience problems and proposed a framework to overcome the same for future studies.