ERD analysis method in motor imagery brain–computer interfaces for accurate switch input

详细信息    查看全文

• 作者：Shuhei Nagamori ; Hisaya Tanaka
• 关键词：BCI ; ERD ; ERS ; Motor imagery ; Mu rhythm
• 刊名：Artificial Life and Robotics
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：22
• 期：1
• 页码：83-89
• 全文大小：
• 刊物类别：Computer Science
• 刊物主题：Artificial Intelligence (incl. Robotics); Computation by Abstract Devices; Control, Robotics, Mechatronics;
• 出版者：Springer Japan
• ISSN：1614-7456
• 卷排序：22

文摘

Motor imagery brain–computer interface (MIBCI) can control computers using MI. However, input accuracy is low, partly owing to individual variability in event-related desynchronization (ERD) detection among different subjects. In an earlier study, we determined that using a max power in the mu-band method, i.e., the peak trace method (PTM), is effective for ERD detection. In this study, we compare the PTM to the band power method to determine the most effective method for ERD detection during MI tasks. Experimental results indicate that we could detect ERD using the PTM; however, MI-state estimation was difficult. We also found that the PTM might be effective for ERD detection in subjects with MI experience.