摘要
设计了基于左、右手运动想象的脑电信号预处理、共同空域模式特征提取、SVM分类在线算法,开发了无线发射、接收开关硬件模块,实现了在线脑电开关系统。受试者可以用脑电波来遥控电灯的关开,这为重症瘫痪病人拓展其与自然的直接交流开辟了新的通道。5位健康的受试者参与了训练实验和在线实验,实验结果表明:经过特定训练,受试者均可有效控制该脑电开关系统,其平均正确率达90%,单个指令输出时间平均为4s。
An online motor imagery detect algorithm is proposed, including electroencephalo- graphy (EEG) preprocessing, common spatial patterns' feature extraction, support vector ma- chine (SVM) classification, and threshold mechanism. By designing the light wireless switch hardware module, a light switch system controlled by EEG is developed. Participants can use their brain waves to remotely control turn-on and turn-off of the light, which especially pro- vides a new direct communication channel for patients suffering from amyotrophic lateral scle- rosis (ALS) with the nature. Five healthy subjects participate in a training experiment and an online experiment. The experimental results show that, after a specific training procedure, participants can effectively control the light switch system with an average accuracy of 90 % and an average time of 4 s for each single command.
出处
《数据采集与处理》
CSCD
北大核心
2013年第6期828-833,共6页
Journal of Data Acquisition and Processing
基金
国家高科技发展研究计划("八六三"计划)(2012AA011601)资助项目
广东高校优秀青年创新人才培养计划(LYM 11122)资助项目
2012年度江门市第一批产业技术研究与开发项目(江财工[2012]156号)资助项目
广东省大学生创新创业训练(1134912036)资助项目
智能机器人湖北省重点实验室开放基金(HBIR200904)资助项目
五邑大学青年科研基金(2013zk08)资助项目
关键词
脑机接口
运动想象
异步系统
共空域模式
motor imagery
brain-computer interface' asynchronous system
common spatial patterns