电子学报 ›› 2020, Vol. 48 ›› Issue (12): 2469-2475.DOI: 10.3969/j.issn.0372-2112.2020.12.024

• 学术论文 • 上一篇    下一篇

基于变分模态分解的癫痫脑电信号分类方法

张学军1,2, 景鹏1, 何涛1,2, 孙知信3,4   

  1. 1. 南京邮电大学电子与光学工程学院, 江苏南京 210023;
    2. 南京邮电大学射频集成与微组装技术国家地方联合工程实验室, 江苏南京 210023;
    3. 南京邮电大学江苏省邮政大数据技术与应用工程研究中心, 江苏南京 210003;
    4. 南京邮电大学国家邮政局邮政行业技术研发中心(物联网技术), 江苏南京 210003
  • 收稿日期:2019-11-25 修回日期:2020-03-21 出版日期:2020-12-25 发布日期:2020-12-25
  • 作者简介:张学军 男,1969年生于江苏南通.工学博士.现为南京邮电大学教授,硕士生导师.研究方向为智能信息处理、认知网络频谱感知、深度学习等.E-mail:zhxj@njupt.edu.cn;景鹏 男,1995年生于江苏泰州.硕士研究生.研究方向为智能信息处理、脑机接口技术.E-mail:japee@foxmail.com
  • 基金资助:
    国家自然科学基金(No.61972208,No.61672299)

An Epileptic Electroencephalogram Signal Classification Method Based on Variational Mode Decomposition

ZHANG Xue-jun1,2, JING Peng1, HE Tao1,2, SUN Zhi-xin3,4   

  1. 1. School of Electronic and Optical Engineering, Nanjing University of Posts and Telecommunications, Nanjing, Jiangsu 210023, China;
    2. Nation-Local Joint Project Engineering Lab of RF Integration&Micropackage, Nanjing University of Posts and Telecommunications, Nanjing, Jiangsu 210023, China;
    3. Post Big Data Technology and Application Engineering Research Center of Jiangsu Province, Nanjing University of Posts and Telecommunications, Nanjing, Jiangsu 210003, China;
    4. Post Industry Technology Research and Development Center of the State Posts Bureau(Internet of Things Technology), Nanjing University of Posts and Telecommunications, Nanjing, Jiangsu 210003, China
  • Received:2019-11-25 Revised:2020-03-21 Online:2020-12-25 Published:2020-12-25

摘要: 癫痫是一种常见的脑部疾病,通过脑电图能非侵入地定位人脑中的致痫区域.为了辨别病灶性和非病灶性癫痫脑电信号,文章提出一种基于变分模态分解的癫痫脑电信号自动检测方法,首先将原信号分割成多个子信号,并对各子信号进行变分模态分解,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和精细复合多尺度模糊熵两个特征并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94.24%,95.58%和90.64%,ROC曲线下面积达0.978.

关键词: 癫痫脑电, 变分模态分解, 精细复合多尺度散布熵, 精细复合多尺度模糊熵, 支持向量机

Abstract: Epilepsy is a recurrent cerebral disease,and electroencephalogram (EEG) provides a non-invasive way to identify epileptogenic sites in the brain.In order to distinguish focal and non-focal epilepsy EEG signals,this paper proposes an automated epileptic EEG detection method based on variational mode decomposition.Firstly,the original signals are divided into several sub-signals,which are decomposed into intrinsic mode functions by using the variational mode decomposition (VMD).Furthermore,refined composite multiscale dispersion entropy (RCMDE) and refined composite multiscale fuzzy entropy (RCMFE) are extracted from each intrinsic mode function.Finally,the support vector machine (SVM) is used to classify characteristics.For an epilepsy EEG signals' public data set,the final experimental performance measures of accuracy,sensitivity,and specificity reach 94.24%,95.58% and 90.64% respectively,and the area under the ROC curve is 0.978.

Key words: epileptic electroencephalogram, variational mode decomposition, refined composite multiscale dispersion entropy, refined composite multiscale fuzzy entropy, support vector machine

中图分类号: