Development of an adaptive BMI for cooperated predictive control by brain-contact

点击翻译按钮获取中文摘要

Brain-machine interface(BMI) has been as a communication tool of a person who lost body function. Extracting of functional information from brain signals is important for controlling a BMI in a realistic and natural way. In this study, we measured functional near-infrared spectroscopy(fNIRS) signals during isometric arm movements in four directions, and performed direction classification. As a result, the classification accuracy was highest when both of OxyHb and DeoxyHb was used as the features and both of time and channel was selected. The peak time of the signal, the task ends, and after a few seconds of task ends, was particularly well chosen.