基于感兴趣脑区LASSO-Granger因果关系的脑电特征提取算法

Feature Extraction of Electroencephalography Based on LASSO-Granger Causality Between Brain Region of Interest

该文将脑功能网络引入到脑电特征提取的研究中,提出一种基于感兴趣脑区LASSO-Granger因果关系的新方法,克服了当前基于孤立脑区的研究方法的不足。先利用主成分分析提取各感兴趣区的最大主成分,然后计算它们之间的LASSO-Granger因果度量,并将其作为特征向量,最后输入支持向量机分类器,对BCI Competition IV dataset 1中的4组数据进行分类识别。结果表明,基于感兴趣脑区间LASSO-Granger因果关系分析和支持向量机分类器的方法对不同的运动想象任务识别率较高,提供了新的研究思路。

Brain functional network is introduced to feature extraction of Electro Encephalo Graphy（EEG）, and a novel method is proposed based on Least Absolute Shrinkage and Selection Operator（LASSO）-Granger causality between Region Of Interest（ROI） in the brain, in order to overcome the inherent deficiencies of research methods based on isolated brain region. Firstly, the maximum principal component of ROIs is extracted by Principal Component Analysis（PCA）, and then causality values between ROIs are calculated by LASSO-Granger. Finally, the values are used as the input vector for Support Vector Machine（SVM）, and then four datasets of BCI Competition IV Dataset 1 are used for classification.Experimental results show that different motor imagery tasks are successfully identified by the method of SVM classifier combined with feature extraction which is based on LASSO-Granger causality between the brain region of interest（ROIs）. This method provides a new idea for the study of extracting EEG features.