文献类型：期刊文献

英文题名：A Dual Stimuli Approach Combined with Convolutional Neural Network to Improve Information Transfer Rate of Event-Related Potential-Based Brain-Computer Interface

作者：Li, Wei[1];Li, Mengfan[2];Zhou, Huihui[3];Chen, Genshe[4];Jin, Jing[5];Duan, Feng[6]

机构：[1]Calif State Univ Bakersfield, Dept Comp & Elect Engn & Comp Sci, Bakersfield, CA 93311 USA;[2]Hebei Univ Technol, State Key Lab Reliabil & Intelligence Elect Equip, Tianjin 300401, Peoples R China;[3]Chinese Acad Sci, Shenzhen Inst Adv Technol, Shenzhen 518055, Guangdong, Peoples R China;[4]Intelligent Fus Technol, Germantown, MD 41061 USA;[5]East China Univ Sci & Technol, Shanghai 200237, Peoples R China;[6]Nankai Univ, Coll Comp & Control Engn, Tianjin 300071, Peoples R China

年份：2018

卷号：28

期号：10

外文期刊名：INTERNATIONAL JOURNAL OF NEURAL SYSTEMS

收录：;EI(收录号：20183705804933);WOS:【SSCI(收录号:WOS:000453425900005)，SCI-EXPANDED(收录号:WOS:000453425900005)】；

基金：The authors would like to thank Mr. Xiaoqian Mao for his help in the experiments conducted at Tianjin University. Mengfan Li was partially funded by the National Natural Science Foundation of China (No. 61473207). This work was also partially funded by CAS Hundred Talent program and grant (No. 172644KYSB20160175), Shenzhen grant (No. JCYJ 20151030140325151, GJHZ20160229200136090, KQ TD20140630180249366).

语种：英文

外文关键词：Dual stimuli; N200; P300; convolutional neural network; information transfer rate

摘要：Increasing command generation rate of an event-related potential-based brain-robot system is challenging, because of limited information transfer rate of a brain-computer interface system. To improve the rate, we propose a dual stimuli approach that is flashing a robot image and is scanning another robot image simultaneously. Two kinds of event-related potentials, N200 and P300 potentials, evoked in this dual stimuli condition are decoded by a convolutional neural network. Compared with the traditional approaches, this proposed approach significantly improves the online information transfer rate from 23.0 or 17.8 to 39.1 bits/min at an accuracy of 91.7%. These results suggest that combining multiple types of stimuli to evoke distinguishable ERPs might be a promising direction to improve the command generation rate in the brain-computer interface.