Classification of Hand Movement Stages for Brain–Computer Interface Using Convolutional Neural Network

In this paper, a deep learning based control algorithm for a fully functional, real-time prosthetic limb is presented to provide motor rehabilitation and restoration. The proposed scheme detects six different events related to hand movement during a task of grasping and lifting an object, using electroencephalographs (EEG). These six events represent the six sequential stages of a grasp-and-lift action like hand starts moving, starts lifting the object. This method mainly aims at increasing the classification performance of BCI applications by incorporating convolutional neural network (CNN) and low-pass filtering based learning approaches on electroencephalography recordings.