基于序贯似然比检验的运动想象脑电信号分类方法研究

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,〖JP2〗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,〖JP〗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．

SPRT-based classification method for motor imagery electroencephalogram

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.