Brain Computer Interface (BCI)
Recent developments in Brain Computer Interface (BCI) technology potentially open a window to allowing our brain to directly communicate with the outside world without the use of muscles. These developments can potentially bring independence and a high quality of life to millions of individuals who have mobility impairment. The main motivation for this research is to develop a new innovative electroencephalography (EEG)-based BCI for communication and control in daily life. BCIs based on Steady State Visual Evoked Potentials (SSVEPs), which are stable oscillations evoked by repetitive visual stimuli, have been identified as ideal for achieving this goal.
The aim of this research is to develop a novel EEG-based BCI to meet the challenges of practical application for conventional assistive devices in the real world, outside the laboratory. The main contributions of the thesis include fundamental scientific research into advanced signal processing, optimization of operation protocols for rehabilitation applications, investigation and identification of impacts for optimizing EEG recording.
Although the canonical correlation analysis (CCA) has been applied extensively and successfully to SSVEP recognition, the spontaneous EEG activities and artifacts that often occur during data recording can deteriorate the recognition performance. Therefore, it is meaningful to extract a few frequency sub-bands of interest to avoid or reduce the influence of unrelated brain activity and artifacts. We studied an improved method to detect the frequency component associated with SSVEP using multivariate empirical mode decomposition (MEMD) and CCA (MEMD-CCA). EEG signals from nine healthy volunteers were recorded to evaluate the performance of the proposed method for SSVEP recognition.
A hybrid multifunctional BCI system that combines brain rhythms (motor imagery) and SSVEP signals has been developed and implemented to verify its performance in a video game. The hybrid BCI system extends the current capabilities and user experience of BCIs in order for it to be implemented to reduce users’ fatigue, which is a critical issue in training sessions in rehabilitation applications. This has also validated the entire design from signal collecting and data processing through to mechanical system simulations and system functionality, which can be diversified by adopting multi-kind BCIs.