Brain waves to motor imagery tasks classification Using machine learning approaches

Abstract

Brain Computer Interfaces (BCI) provides an interface between external devices such as computers for the reading of the human mind, remote communication, and interaction with disabled and paralyzed people. Electroencephalogram (EEG) signals are used for BCI systems. Motor Imagery (MI) based BCI system allows a person to perform motor movements by having the imagination of the movement of the body without performing any movement of the limb or any external stimulus. The person’s intention is translated via the imagination of the motor movement through the brain using EEG signals. An efficient way of communication is required to be designed for speech-impaired persons. So, EEG-based vowel classification is one of the motivations in this direction. The inspiration for EEG-based digit classification is the analysis of EEG signal classification in a complex scenario containing more than five classes of EEG signals with a very large number of its instances using the proposed model.