基于增强深度特征和TSK模糊分类器的癫痫脑电信号识别
CSTR:
作者:
作者单位:

1. 湖州师范学院 信息工程学院,浙江 湖州 313000;2. 浙江省现代农业资源智慧管理与应用研究 重点实验室,浙江 湖州 313000;3. 湖州学院 理工学院,浙江 湖州 313000

作者简介:

通讯作者:

E-mail: 1047897965@qq.com.

中图分类号:

TP273

基金项目:

国家自然科学基金面上项目(61771193,61802123,61772199).


TSK fuzzy classifier based on enhanced deep feature for epilepsy EEG signal recognition
Author:
Affiliation:

1. School of Information Engineering,Huzhou University,Huzhou 313000,China;2. Zhejiang Province Key Laboratory of Smart Management & Application of Modern Agricultural Resources,Huzhou 313000,China;3. School of Science and Engineering,Huzhou College,Huzhou 313000,China

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    识别癫痫脑电信号的关键在于获取有效的特征和构建可解释的分类器.为此,提出一种基于增强深度特征的TSK模糊分类器(ED-TSK-FC).首先,ED-TSK-FC使用一维卷积神经网络(1D-CNN)自动获取癫痫脑电信号的深度特征与潜在类别信息,并将深度特征和潜在类别信息合并为增强深度特征;其次,将增强深度特征作为ED-TSK-FC模糊规则前件与后件部分的训练变量,保证原始输入的深度特征及其潜在意义都出现在模糊规则中,进而对增强深度特征作出良好的解释;然后,采用岭回归极限学习算法对模糊规则的后件参数进行快速求解,在不显著降低分类准确度的情况下,ED-TSK-FC的廉价训练方法可以缩短模型的训练时间;最后,在Bonn癫痫数据集上,分别从分类性能、学习效率和可解释性3个方面,验证ED-TSK-FC的优越性.

    Abstract:

    The key to recognize epileptic EEG signals is to obtain effective features and construct an interpretable classifier. In this paper, a Takagi-Sugeno-Kang (TSK) fuzzy classifier based on enhanced deep feature (ED-TSK-FC) for epilepsy EEG signal recognition is proposed. The ED-TSK-FC adopts an one-dimension convolution neural network (1D-CNN) to extract deep feature and label information, label information is expanded to the deep feature space to generate the enhanced deep feature. As the antecedent-consequent variable of the ED-TSK-FC, the enhanced deep feature can provide deep feature and label information for each fuzzy rule, hence fuzzy rules open the black box of the 1D-CNN in an interpretable manner. Finally, the ridge regression extreme learning algorithm is used to quickly solve the consequent parameters of fuzzy rules. In addition, the ED-TSK-FC also formulates a cheap strategy to decrease training time. Experiments show that this method provides an excellent performance on the Bonn epilepsy dataset in terms of classification performance, training time and interpretability.

    参考文献
    相似文献
    引证文献
引用本文

蒋云良,翁江玮,申情,等.基于增强深度特征和TSK模糊分类器的癫痫脑电信号识别[J].控制与决策,2023,38(1):171-180

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2022-12-23
  • 出版日期: 2023-01-20
文章二维码