摘要
在左右手运动想象的脑电(EEG)分析方法中,目前大多针对多通道采集的EEG数据,难以应用到单通道脑机接口(BCI)中。本文采用一种改进的独立成分分析(ICA)方法,实现了对EEG数据有效的预处理。首先,对数据进行线性漂移校正,去除数据漂移;然后采用延时窗口数据增加虚拟通道数,利用ICA除去EEG数据中的伪迹,即眼电和心电;最后利用希尔伯特-黄变换(HHT)后的瞬时幅值,求平均瞬时能量特征并分类。实验证明,该方法测试性完成了EEG数据的预处理工作,提高了单通道EEG信号的分类率,可为单通道的便携式BCI研究打下基础。
Most of electroencephalogram （EEG） acquired by multi-channels is difficult to be applied to the single- channel brain-computer interface （BCI） in the EEG analysis method based on left and right hand motor imagery. The present research applied an improved independent component analysis （ICA） method to realize pretreatment of the EEG effectively. Firstly, data drift was removed through linear drift correction. Secondly, the number of virtual channels were increased by applying delayed window data and some EEG artifacts which are namely electrooculogram （EOG） and electrocardiogram （ECG） were removed by ICA. Finally, the average instantaneous energy characteristics were calculated and classified through the instantaneous amplitude which was solved by applying Hilbert-Huang transform （HHT）. The experiment proves that the method completes the EEG pretreatment and improves classifica- tion ratio of single-channel EEG, and lays a foundation of single-channel and portable BCI.
作者
李松
伏云发
杨秋红
刘传伟
孙会文
LI Song FU Yunfa YANG Qiuhong LIU Chuanwei SUN Huiwen(School of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, Chin)
出处
《生物医学工程学杂志》
EI
CAS
CSCD
北大核心
2016年第5期862-866,共5页
Journal of Biomedical Engineering
基金
国家自然科学基金资助项目(81470084
61463024)
云南省应用基础研究计划资助项目(2013FB026)
云南省级人培资助项目(KKSY201303048)
云南省教育厅重点资助项目(2013Z130)
昆明理工大学脑信息处理与脑机交互融合控制(学科方向团队建设经费)资助项目
关键词
单通道
独立成分分析
延时窗口
虚拟通道
运动想象
single channel
independent component analysis
delayed window
virtual channel
motor imagery
