《计算机应用研究》|Application Research of Computers

基于跨层全连接神经网络的癫痫发作期识别

Epileptic EEG identification with cross layer fully connected neural network

免费全文下载 (已被下载 次)  
获取PDF全文
作者 王凤琴,卢官明,柯亨进,肖新凤
机构 1.湖北师范大学 物理与电子科学学院,湖北 黄石 435102;2.南京邮电大学 信息与通信工程学院,南京 210003;3.武汉大学 计算机学院,武汉 430072;4.广东环境保护工程职业学院,广州 528216
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2019)07-039-2098-06
DOI 10.19734/j.issn.1001-3695.2018.01.0017
摘要 在缺乏足够先验知识下,自适应癫痫发作期识别异常困难。提出一种新的度量通道之间的同步特征计算方法(聚类划分互信息),以相关矩阵方式组织单窗口内全局同步特征模式,进而设计一种跨层全连接神经网络分类器,对非平稳同步特征模式实现自适应分类。实验表明该方法可获得[98.19%±0.24%]精确度,[98.27%±0.51%]敏感度和[98.11%±0.36%]特异度,超过了现有大部分方法的分类性能。另外,该方法无须去噪和去伪迹等预处理过程;而且其仅需设置一个超参数(时间窗),避免了过多的潜在错误参数设置而导致的分类性能的降低。
关键词 聚类划分互信息; 脑电; 癫痫; 同步; 模式分类; 跨层全连接神经网络
基金项目 国家自然科学基金资助项目(61071167,61501249)
本文URL http://www.arocmag.com/article/01-2019-07-039.html
英文标题 Epileptic EEG identification with cross layer fully connected neural network
作者英文名 Wang Fengqin, Lu Guanming, Ke Hengjin, Xiao Xinfeng
机构英文名 1.College of Physics & Electronic Science,Hubei Normal University,Huangshi Hubei 435102,China;2.College of Telecommunications & Information Engineering,Nanjing University of Posts & Telecommunications,Nanjing 210003,China;3.School of Computer Science,Wuhan University,Wuhan 430072,China;4.Guangdong Polytechnic of Environmental Protection Engineering,Guangzhou 528216,China
英文摘要 Under the circumstance of insufficient prior knowledge, it becomes even more important to adaptively classify the synchronization dynamics to accurately characterize the intrinsic nature of seizure activities represented by the EEG. This paper first measured the global synchronization by calculating clustering partition mutual information(MI) of all EEG data channels. Then it designed a cross layer fully connected net to adaptively characterize the synchronization dynamics captured correlation matrices and automatically identify the seizure states of the EEG. It also performed experiments over the CHB-MIT scalp EEG dataset to evaluate the proposed approach. It identified seizure states with an accuracy, sensitivity and specificity of [98.19%±0.24%], [98.27%±0.51%], and[98.11%±0.36%], respectively. The resulted performance was superior to those of most existing methods over the same dataset. The approach alleviated the need for strictly denoising and artifact removing based on the EEG prior knowledge that is mandatory for existing methods. Only one hyper-parameter need be set manually to avoid getting worse performance because of complex parameter setting.
英文关键词 clustering partition mutual information; EEG; epilepsy; synchronization; pattern classification; cross layer fully connected net
参考文献 查看稿件参考文献
 
收稿日期 2018/1/5
修回日期 2018/3/3
页码 2098-2103
中图分类号 TP391
文献标志码 A