Speaker
Description
One of the studies of the active brain-computer interface (BCI) focuses on identifying movements from human neurophysiological signals to control external devices such as robotic arms. In the literature, EEG-based BCI is utilized to decode user's information to perform action or fill the gap from the brain to the arms in the case of illness. The purpose of this work is to understand, among the scientific literature, what the best Deep Learning (DL) architecture is for motor execution (ME) classification. Data from 105 people from the Physionet dataset and 15 subjects from the Upper Limb dataset were used. EEGnetv4, Deep4Net, and EEGITnet were used to classify EEG signals under ME for real-time BCI. A comparison of three different types of DL architecture training the network from scratch, using the data from the Physionet and Upper Limb, preprocessed in the same way to make them similar, was provided using filtering, ICA decomposition and algorithm for automatic artifacts removal. In this way, the comparison of results is not influenced by the bias introduced in each step of feature cleaning, extraction, or selection. To the best of our knowledge, no other studies have conducted a similar analysis, making it impossible to compare this achievement with others previously published in the literature. The best results were achieved from the EEGnetv4 trained without Common Spatial Pattern transformation: 0.70 and 0.77 of accuracy for Physionet and Upper Limb dataset respectively.
