融合注意力机制的多视图卷积网络癫痫智能辅助检测  

Multi-view Convolutional Network with Fused Attention for Intelligent-assisted Epilepsy Detection and Recognition

作　　者：李奇[1,2] 闫旭荣 武岩[1,2] 赵迪 常立娜 孙瀚琳 LI Qi;YAN Xu-rong;WU Yan;ZHAO Di;CHANG Li-na;SUN Han-lin(School of Computer Science and Technology,Changchun University of Science and Technology,Changchun 130022,China;Zhongshan Institute,Changchun University of Science and Technology,Zhongshan 528400,China)

机构地区：[1]长春理工大学计算机科学技术学院,长春130022 [2]长春理工大学中山研究院,中山528400

出　　处：《科学技术与工程》2025年第5期1988-1995,共8页Science Technology and Engineering

基　　金：吉林省科技发展计划国际科技合作项目(20200801035GH);吉林省科技发展计划国际联合研究中心建设项目(20200802004GH)。

摘　　要：针对单一视图网络癫痫检测识别精度低的问题,提出一种融合注意力机制的多视图卷积网络癫痫智能辅助检测模型(multi-view convolutional network with fused attention mechanism,FAM-MCNN)。该模型从时域、频域、时频域和非线性域提取多视图特征来全面表征脑电信号;采用多尺度卷积捕捉不同层次的细节信息;引入注意力机制分别从视图维度和单个特征向量维度对特征进行加权融合,从而提高对癫痫患者不同类别脑电信号的区分能力。在CHB-MIT癫痫数据集上进行的对比实验结果显示,与单一视图网络相比,FAM-MCNN模型的平均准确率、灵敏度、特异度分别提高了14.29%、16.13%、12.54%。此外,对该模型采用少量训练样本(25%)进行实验,结果显示其检测性能达到了拥有大量训练样本(80%~90%)的对比模型水平。In response to the problem of low accuracy in epilepsy detection and recognition using single-view networks,a multi-view convolutional network model with fused attention mechanism(FAM-MCNN)was proposed.Multiple view features were extracted from time domain,frequency domain,time-frequency domain and nonlinear domain to characterize electroencephalogram(EEG)signals comprehensively.Multi-scale convolution was used to capture different levels of detail information.In order to improve the ability to distinguish different types of EEG signals in epileptic patients,the attention mechanism was introduced to combine the features from view dimension and single feature vector dimension respectively.The results of the comparison experiments performed on the CHB-MIT epilepsy dataset show that the average accuracy,sensitivity,and specificity of the FAM-MCNN model are improved by 14.29%,16.13%,and 12.54%,respectively,when compared to a single-view network.In addition,experiments under a small number of training samples(25%)show that its detection performance reaches the level of the comparison model with a large number of training samples(80%~90%).

关 键 词：脑电信号 多视图卷积 注意力机制 癫痫智能辅助检测 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]