下肢想象动作的脑电特征提取是基于脑-机接口技术实现下肢运动神经重建的关键点和难点。通过以下肢运动神经重建的首个关键动作-站起为切入点,设计了相应的以复合动作诱发电位为基础的站起想象动作脑电采集新方案,而后采用扩展最大熵独立分量分析(extended informax ICA)结合连续小波功率谱分析的新方法提取大脑在站起想象动作时产生的事件相关同步/去同步(ERD/ERS)信息。经实验数据分析证明上述方法行之有效,可以为脑-机接口技术在下肢运动重建中的应用提供有效的手段。
EEG feature extraction from imaginary lower limb motor tasks is the key and difficult point of motor nerve reconstruction for lower limbs based on brain-computer interface technology. In this paper, stand-up performance, which is the first important action for lower limb motor nerve reconstruction, is selected as a starting point for the research work, and the corresponding EEG acquisition program for imaginary stand-up tasks is designed based on evoked compound action potential. Then a new method based on extended Informax ICA combined with continuous wavelet power spectral analysis is presented to extract the information of event related desynchronization or event related synchronization (ERD/ERS) produced by imaginary standing up in brain. The results deduced from experimental data analysis show that the above mentioned method is proved to be effective, and can provide an effective way for the application of brain-computer interface in lower limb motor nerve reconstruction.