Study of Distinction of electroencephalogram via steady state visual evoked potentioal under the condition of luival blinking animations for visual stimulus
When the disease called Amyotrophic Lateral Sclerosis (ALS) had progressed, the muscular strength of patient decreases. Then, patients cannot control a wheelchair by their own ability. In this case, Brain Machine Interface (BMI) helps such patients to control the electric wheelchair by their own intention. BMI enables a human to control a machine using the electroencephalogram (EEG). In this study, a detection method of steady state visual evoked potential(SSVEP) in short time was investigated for the sake of future use in BMI. In addition, experiments for distinguishing SSVEP were examined under the condition that plural visual stimuli were presented simultaneously. Preliminary experiments were carried out to determine the form and color of a blinking symbol and the number of the electrodes to be used. As results, the blinking symbol of 10Hz and 12Hz was chosen as white square with black background, and the blinking symbol of 15Hz was chosen as white square with blue background. In addition, three electrodes were chosen to be used. Using the short time Fourier transform(STFT) and the moving-average method, the analysis method of electroencephalogram (EEG) for detecting SSVEP and detection conditions of SSVEP were determined. Then, experiments for distinguishing SSVEP were executed under the condition that three kinds of the blinking animation, in which the white square symbol blinks in 10Hz, 12Hz, and 15Hz respectively, were presented simultaneously as visual stimulus. Accuracy rate of more than 90% was obtained for distinction of each stimulus frequency in 2 to 6 sec.