EMD-based Feature Extraction from Motor Imaginary EEG of Complex Movements
In this study,we proposed and evaluated the use of the empirical mode decomposition (EMD) technique to extract feature information of the event-related (de) synchronization (ERD/ERS) phenomenon during complex motor imagination of combined body and limb action.The EEG data were separated into intrinsic mode functions (IMFs) using the EMD method and determined the characteristic IMFs by power spectral density (PSD) analysis.Thereafter,the analytic signals of the characteristic IMFs can be obtained by the Hilbert transformation,then extracting the ERD/ERS feature of each single-trial.To verify the effectiveness of this method,ten subjects were tested for distinguishing three kinds of complex motor imagery.The classification performance suggests that the proposed EMD based approach is effective for ERD/ERS feature extraction and is worth for the further application in a brain-computer interface.
complex motor imagery event-related (de) synchronization empirical mode decomposition intrinsic mode functions brain-computer interface
Weibo Yi Dong Ming Shuang Qiu Minpeng Xu Xingwei An Hongzhi Qi Baikun Wan
Department of Biomedical Engineering, Tianjin University, Tianjin 300072, P. R. China
国际会议
World Congress on Medical Physics and Biomedical Engineering (2012年医学物理及生物医学工程国际会议(IFMBE))
北京
英文
1541-1544
2012-05-26（万方平台首次上网日期，不代表论文的发表时间）