Decoding Movement Direction for Brain-Computer Interfaces using Depth and Surface EEG Recordings

A Brain-Computer Interface (BCI) is a communication system between the brain and an external device. The purpose of this project was to develop a decoder for a BCI system capable of providing a control output based on decoding of the intention of movement and of different directions of movement execution, which in turn will enhance the quality of the command to external systems to propitiate the restoration of more complex motor functions than the two-choice commands commonly available in literature. The system was based on classification of invasive and non-invasive brain signal recordings. Results of the detection of the intention of movement and the classification of the direction were significantly above the level of chance for both iEEG and scalp EEG data. The present study also enabled to set a comparison of different methods used for spatial filtering, normalization and classification