面向康复工程的脑电信号分析和判别方法研究

Research on EEG Signal Analysis and Discrimination Method For Rehabilitation Engineering

为提高脑电信号的分析和判别的准确率,利用数据挖掘的方法对数据进行处理,并建立脑电信号判别模型.首先,对P300信号标志符进行识别,完成对脑电信号的数据预处理,利用经验模态分解方法(Empirical Mode Decomposition,EMD)降低噪声去除伪迹,使用主成分分析方法(Principal Component Analysis,PCA)进行特征提取.然后,采用合成少数类过采样技术方法(Synthetic Minority Oversampling Technique,SMOTE)扩充样本使正负样本数量达到均衡.最后,采用支持向量机方法(Support Vector Machine,SVM)建立二分类模型完成对P300信号标识符进行识别.经运行结果可追,SMOTE改进的分类模型结果精度都有所改进,准确性有明显提高,能准确识别P300信号标识符.

In order to improve the accuracy of EEG analysis and discrimination,data mining method is used to process the data,and EEG discrimination model is established.Firstly,P300 signal markers are identified,and the data preprocessing of EEG signals is completed.Empirical Mode Decomposition(EMD)is used to reduce noise and remove artifacts,and Principal Component Analysis(PCA)is used to extract features.Then,the synthetic minority over sampling technique(SMOTE)is used to expand the samples to balance the number of positive and negative samples.At last,Support Vector Machine(SVM)is used to establish a two-class model to complete the identification of P300 signal identifiers.The running results can be traced.SMOTE's improved classification model has improved the accuracy and accuracy,and can accurately identify P300 signal identifiers.