基于多通道脑电特征运动意识任务的分类

Classification of the Mental Tasks Based on Multichannel Electroencephalogram Features

提出利用多通道脑电(EEG)相关性信息作为特征,实现对想象左右手运动两种意识任务的分类.分析了两种意识任务下描述多通道EEG相关性的两个参数空间复杂度和有效场强的变化,发现想象任务开始时,对侧脑区空间复杂度增高,有效场强降低,而同侧脑区的变化恰好相反,较好地反映了想象左右手运动时大脑两侧运动皮层状态的不同.将两种特征联合用于分类,得到了满意的结果,从而为脑-计算机接口系统中意识任务的特征提取与分类提供了新思路.

Multichannel electroencephalogram （EEG） correlation features are extracted to classify left and right hand motor imagery tasks. The time courses of spatial complexity Ω and field strength ∑, are analyzed respectively as two imagery tasks are performed. It is demonstrated that at the onset of imagination,Ω calculated by multichannel EEG from the contralateral and mid-central region rises while the corresponding ∑,lowers. The opposite changes occur with Ω and ∑, from the ipsilateral and mid-central region. Thus, the difference of brain states in two hemispheres during left or right hand motor imagery is well represented.The features Ω and ∑, are combined to discriminate left and right hand motor imagery tasks and satisfactory results are obtained, which can be expected to provide a new way for the classification of mental tasks in brain computer interface.