[19a-Z32-1~10] 23.1 Joint Session N "Informatics"

[19a-Z32-5] Decoding from somatosensory EEG response using machine learning for robot control brain machine interface

〇(B)Daisuke Konno, Riku Takamura, Hirohito Sawahata

Keywords:Event related potential

To develop the brain-machine interface (BMI) for artificial limb control, we focused the tactile sensory electroencephalogram (EEG) changes due to conscious attention to body parts. Using the support vector machine (SVM), we examined whether it was possible to distinguish between the case where only one of the sensory stimuli for the thumb and index finger was consciously paid attention and the case where it was not. As the results, accuracy of 60% was obtained. It was suggest that the tactile sensory EEG signal could be used to the BMI.