基于脑电主成分分析和k-最近邻的多动症儿童与正常儿童分类研究

发布时间：2018-02-28 21:07

本文关键词： 脑电图多动症干扰控制主成分分析分类　出处：《生物医学工程学杂志》2016年02期 　论文类型：期刊论文

【摘要】：本文尝试通过脑电信号检测方法辅助多动症儿童进行临床个体化诊断。首先基于一种经典的干扰控制试验任务Simon-spatial Stroop范例采集14名多动症儿童和16名正常儿童的脑电数据,并完成滤波、分段、去伪迹等预处理;然后采用主成分分析(PCA)进行电极优化选择,分别选取每种刺激模式下出现率90%以上的优化电极作为共有电极,并提取共有电极潜伏期(200~450ms)波幅的均值特征;最后采用基于欧氏距离的k-最近邻(KNN)和基于径向基核函数的支持向量机(SVM)分类器来分类。实验发现同种试验任务中多动症儿童比正常儿童表现出更低的反应正确率和更长的反应时间;多动症儿童与正常儿童的前额叶优化电极均出现N2,顶枕叶均有P2出现,且多动症儿童的峰值更低;在该实验中KNN分类准确率高于SVM分类器,StI刺激模式下KNN分类器的最高分类准确率为89.29%。以上结果说明,干扰控制试验中多动症儿童与正常儿童的前额叶及顶枕叶的脑电信号存在差异,该结果可为多动症个体的脑电信号临床诊断提供一定科学依据。
[Abstract]:This paper attempts to assist children with ADHD with clinical individualized diagnosis by means of EEG signal detection. Firstly, the EEG data of 14 children with ADHD and 16 normal children were collected based on a classical interference control test task, Simon-spatial Stroop. The preprocessing of filtering, segmenting and removing artifacts was completed, and then the electrode was optimized by principal component analysis (PCA), and the optimal electrode with a rate of more than 90% in each stimulus mode was selected as the common electrode, respectively. The mean values of the amplitudes of the shared electrode latencies (200 ~ 450ms) were extracted. Finally, the classification is done by using K- nearest neighbor (KNN) based on Euclidean distance and support vector machine (SVM) classifier based on radial basis function kernel function. It is found that children with ADHD in the same task have lower response than normal children. Rate and longer reaction time; In children with ADHD and normal children, N2 was found in prefrontal lobe, P2 in parietal occipital lobe, and the peak value was lower in ADHD children. In this experiment, the accuracy of KNN classification is higher than that of KNN classifier under SVM stimulus mode, which is 89.29. The EEG signals in prefrontal lobe and parietal occipital lobe of ADHD children and normal children are different in interference control test. The results can provide a scientific basis for the clinical diagnosis of EEG in ADHD individuals.
【作者单位】：常州大学信息科学与工程学院;常州市生物医学信息技术重点实验室;苏州大学附属第三医院脑科学研究中心;
【基金】：国家自然科学基金资助项目(61201096) 常州市科技资助项目(CE20145055) 江苏省“青蓝工程”资助项目
【分类号】：R749.94

【相似文献】