Detecting Motor Imagery Movement from EEG Signal

Abstract

Electroencephalography (EEG) signals can be captured with the help of Brain-Computer Interfaces. When properly analyzed and applied, the information in these EEG signals can serve various purposes. People who are paralyzed or partially paralyzed and have difficulty communicating as a result of their condition can benefit immensely from the use of EEG. By detecting the motor imagery movement from EEG, we can determine the intent of a subject who is unable to perform motor functions (e.g., paralyzed patient) but is imagining them. However, detecting motor movement from EEG signals is challenging since EEG is susceptible to noise. Moreover, the complex relationship between motor activities and EEG data makes it difficult to classify. Deep neural networks excel at comprehending intricate features and executing complex computations. Using the capabilities of deep neural networks, we develop a hybrid neural network model in this paper that can accurately detect motor activity movement from EEG data; our model outperforms the state-of-the-art models and generates a classification accuracy of 98%.