The classification recognition rate of motor imagery is a key factor to improve the performance of brain–computer interface (BCI). Thus, we propose a feature extraction method based on discrete wavelet transform (DWT), empirical mode decomposition (EMD), and approximate entropy. Firstly, the electroencephalogram (EEG) signal is decomposed into a series of narrow band signals with DWT, then the sub-band signal is decomposed with EMD to get a set of stationary time series, which are called intrinsic mode functions (IMFs). Secondly, the appropriate IMFs for signal reconstruction are selected. Thus, the approximate entropy of the reconstructed signal can be obtained as the corresponding feature vector. Finally, support vector machine (SVM) is used to perform the classification. The proposed method solves the problem of wide frequency band coverage during EMD and further improves the classification accuracy of EEG signal motion imaging.

This paper proposes an improved Empirical mode decomposition (EMD) method piecewise cubic hermite interpolation and mirror extension. Firstly, mirror expansion method is used to expand preliminary signal endpoint; then the piecewise cubic Hermite interpolation algorithm is applied to obtain the envelopes. Finally Intrinsic mode functions (IMFs) are obtained through improved EMD (MPC-HEMD). Through the analysis on simulation signals and Electroencephalogram (EEG) signals, we found that the new method was better than the traditional method.

Network analysis based on graph theory has greatly promoted the cognition of the human brain network. A detailed brain network function connection analysis was carried out for the brain of normal human brain and mental illness patients. We studied the Magnetoencephalography (MEG) of 9 normal subjects and 9 schizophrenics in left hemisphere temporal lobe and frontal lobe regions. And obtained the dynamic function connectivity matrix by calculating Pearson correlation coefficients that based on sliding time window and shorttime Fourier transform, and constructed weight and binary network by graph theory. Analyzed the small world properties of normal human brain networks, and compared the differences of network between normal subjects and patients with schizophrenia.

癫痫脑电信号分类对于癫痫诊治具有重要意义. 为了实现病灶性与非病灶性癫痫脑电信号的分类, 本文利用弹性网回归重构变分模态分解算法, 提出弹性变分模态分解算法并将其应用到所提癫痫脑电信号分类方法中. 该方法先将原信号分割成多个子信号, 并对各子信号进行弹性变分模态分解, 然后从分解后的不同变分模态函数中提取精细复合多尺度散布熵作为特征, 最后利用支持向量机进行分类. 针对癫痫脑电的公共数据集, 最终的实验结果表明, 准确率、灵敏度和特异度三个性能指标分别达到92.54%, 93.22%和91.86%.