基于序贯似然比检验的运动想象脑电信号分类方法研究  

SPRT-based classification method for motor imagery electroencephalogram

作　　者：刘蓉[1] 李春月[1] 王永轩[2] 王媛媛[1] 李响[3] 

机构地区：[1]大连理工大学电子信息与电气工程学部,辽宁大连116024 [2]大连大学附属中山医院,辽宁大连116001 [3]大连医科大学附属第二医院,辽宁大连116027

出　　处：《大连理工大学学报》2013年第6期898-902,共5页Journal of Dalian University of Technology

基　　金：国家自然科学基金资助项目(61005088);"十一五"国家科技支撑计划资助项目(2012BAJ18B06);中央高校基本科研业务费专项资金资助项目(DUT13JB02)

摘　　要：快速准确地对脑电信号进行特征分类是脑-机接口研究的关键问题之一.从人脑决策模型出发,结合自适应小波基特征提取方法,提出了一种基于序贯似然比检验的运动想象脑电信号动态分类方法.该方法在分类中无须预先固定样本量,而是逐次取样,累积分类信息,有利于解决脑-机接口的实时控制问题.为了更好地衡量该方法的有效性,进行了10次10折交叉验证,实验结果表明3个运动想象数据集共8位受试者的平均正确率达到87%以上,互信息和分类时间等指标也表明该方法能够有效提高脑-机接口系统的性能,具有较好的实用性.To extract and classify the electroencephalogram （EEG） signal features fast and accurately is a key issue for the brain-computer interface （BCI） systems. Based on the human decision-making model, a motor imagery EEG dynamic classification method is proposed based on the sequential probability ratio testing （SPRT） combined with an adaptive wavelet feature extraction method. Without pre-fixed sample size, this classification method samples and accumulates classification information successively. It is helpful to solve the real-time control problems in the BCI. In order to evaluate the effectiveness of the algorithm, a 10 times 10-fold cross-validation is used. The experimental results show that the average classification accuracy of three motor imagery datasets of eight subjects is above 87%. The results of mutual information and classification time also show that the method can effectively improve the performance of BCI system and has good practicability.

关 键 词：脑-机接口 运动想象 自适应特征提取 动态分类 

分 类 号：R318[医药卫生—生物医学工程]