Multi-Functional Brain Computer Interface Using Convolutional Neural Networks

Brain-computer interface (BCI) is a promising technology that controls computers or machines using brain signals. With this technology, people with various disabilities, such as neural paralysis, and spinal cord injury can control electric devices or express their intention by thinking. However, previous BCI studies have a limitation that they can predict only one type of intention. To use the BCI system in daily life, the BCI user should be able to achieve various tasks such as moving, text typing, and arm movements. In this paper, we propose a multi-functional BCI method that can predict various intentions simultaneously. To classify multiple intentions, we proposed two prediction models using Neural Networks (NN) and Convolutional Neural Networks (CNN) models. To evaluate the proposed BCI system, the classification accuracy of the model was measured and compared using steady state visually evoked potential (SSVEP), sensory motor rhythm (SMR), and both of them (Multiple Intention). The average prediction accuracies were 22.46% in NN, 55.86% in CNN. These results indicate that the proposed multi-functional BCI can predict multiple intentions. It also means that users of the proposed BCI system can control various electric devices simultaneously.