[1]李红利,丁满,张荣华,等.基于特征融合神经网络的运动想象脑电分类算法[J].中国医学物理学杂志,2022,39(1):69-75.[doi:DOI:10.3969/j.issn.1005-202X.2022.01.012]
　LI?ongli,ING Man,HANG?onghua,et al.Motor imagery EEG classification algorithm based on feature fusion neural network[J].Chinese Journal of Medical Physics,2022,39(1):69-75.[doi:DOI:10.3969/j.issn.1005-202X.2022.01.012]
点击复制

基于特征融合神经网络的运动想象脑电分类算法()
分享到：

《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]

卷:
39卷
期数:
2022年第1期
页码:
69-75
栏目:
医学信号处理与医学仪器
出版日期:
2022-01-17

文章信息/Info

Title:
Motor imagery EEG classification algorithm based on feature fusion neural network
文章编号:
1005-202X（2022）01-0069-07
作者:
李红利1; 丁满1; 张荣华2; 修春波1; 马欣3
1.天津工业大学控制科学与工程学院， 天津 300387； 2.天津工业大学人工智能学院， 天津 300387； 3.天津工业大学电子与信息工程学院， 天津 300387
Author(s):
LI?ongli1; ?ING Man1; ?HANG?onghua2; ?IU?hunbo1;  MA Xin3
关键词:
运动想象; 脑电分类; 神经网络; 特征融合
Keywords:
Keywords: motor imagery electroencephalogram classification neural network feature fusion
分类号:
DOI:10.3969/j.issn.1005-202X.2022.01.012
DOI:
DOI:10.3969/j.issn.1005-202X.2022.01.012
文献标志码:
A
摘要:
运动想象-脑机接口（MI-BCI）技术为运动障碍患者提供了一种新的与外界交流的能力。应用卷积神经网络（CNN）处理运动想象（MI）脑电分类问题时，多提取最后卷积层的特征，忽视了中间层大量可用信息，导致MI-BCI的分类性能较差。针对这一问题，提出模型内层融合（WMFF）和模型间层融合（CMFF）两种特征融合策略。WMFF策略提取CNN每一层特征进行融合；CMFF策略融合CNN和长短时记忆网络并提取每一层特征。本研究用BCI竞赛IV Datasets 2a数据集对所提方法进行验证，WMFF和CMFF MI脑电信号四分类平均正确率分别达到76.19%和80.46%。结果表明，所提方法可有效提高MI脑电信号分类正确率，为MI脑电信号分类提供了新的思路。
Abstract:
Abstract: Motor imagery-based brain computer interface (MI-BCI) technology enables patients with movement disorders to acquire a new ability to communicate with the outside world. However, when using convolutional neural network (CNN) for MI electroencephalogram (EEG) classification, researchers often extract the features of the final convolutional layer and ignore the large amount of available information in the middle layer, resulting in poor classification performance of MI-BCI. To solve this problem, two kinds of feature fusion strategies, namely with-in model fusion-feature (WMFF) and cross model fusion-feature (CMFF), are proposed. WMFF strategy extracts the features of each CNN layer separately for feature fusion while CMFF strategy integrates CNN and long short-term memory network and extracts the features of each layer. BCI competition IV Datasets 2a is used to verify the proposed method, and the results show that the average accuracies of WMFF and CMFF for 4-category MI EEG classification reach 76.19% and 80.46%, respectively, which indicates that the proposed method can effectively improve the accuracy of MI EEG classification, and provide new ideas and methods for the application of MI-BCI.

相似文献/References:

[1]苌文清,孙曜. 基于多变量格兰杰因果关系的运动想象因效网络构建[J].中国医学物理学杂志,2018,35(12):1457.[doi:DOI:10.3969/j.issn.1005-202X.2018.12.017]
　CHANG Wenqing,SUN Yao. Construction of motion imagination causal network based on multivariable Granger causality[J].Chinese Journal of Medical Physics,2018,35(1):1457.[doi:DOI:10.3969/j.issn.1005-202X.2018.12.017]
[2]周杰,杨国雨,徐涛. 基于空间频率与时间序列信息的多类运动想象脑电分类[J].中国医学物理学杂志,2019,36(1):81.[doi:DOI:10.3969/j.issn.1005-202X.2019.01.016]
　ZHOU Jie,YANG Guoyu,XU Tao. Classification of multi-class motor imagery EEG data based on spatial frequency and time-series information[J].Chinese Journal of Medical Physics,2019,36(1):81.[doi:DOI:10.3969/j.issn.1005-202X.2019.01.016]
[3]姜月,邹任玲. 基于多特征融合的运动想象脑电信号识别研究[J].中国医学物理学杂志,2019,36(5):590.[doi:DOI:10.3969/j.issn.1005-202X.2019.05.019]
　JIANG Yue,ZOU Renling. Recognition of motor imagery EEG signals based on multi-feature fusion[J].Chinese Journal of Medical Physics,2019,36(1):590.[doi:DOI:10.3969/j.issn.1005-202X.2019.05.019]
[4]吴拾瑶,随力,杨兰,等.运动想象重塑脑功能的研究进展[J].中国医学物理学杂志,2021,38(11):1449.[doi:DOI:10.3969/j.issn.1005-202X.2021.11.023]
　WU Shiyao,SUI Li,YANG Lan,et al.Research advances in motor imagery for remodeling brain functions[J].Chinese Journal of Medical Physics,2021,38(1):1449.[doi:DOI:10.3969/j.issn.1005-202X.2021.11.023]
[5]贾婷婷,董朝轶,马爽,等.基于互信息特征提取的运动想象脑机接口[J].中国医学物理学杂志,2022,39(1):63.[doi:DOI:10.3969/j.issn.1005-202X.2022.01.011]
　JIA Tingting,DONG Chaoyi,et al.Brain-computer interface of motion imagery based on mutual information-based feature extraction[J].Chinese Journal of Medical Physics,2022,39(1):63.[doi:DOI:10.3969/j.issn.1005-202X.2022.01.011]
[6]戴亮宙,王娆芬,王海玲.基于特征融合AEBGNet的运动想象脑电分类算法[J].中国医学物理学杂志,2024,41(8):1021.[doi:DOI:10.3969/j.issn.1005-202X.2024.08.016]
　DAI Liangzhou,WANG Raofen,WANG Hailing.Motor imagery EEG classification algorithm using feature fusion based AEBGNet[J].Chinese Journal of Medical Physics,2024,41(1):1021.[doi:DOI:10.3969/j.issn.1005-202X.2024.08.016]

备注/Memo

备注/Memo:
【收稿日期】2021-08-10 【基金项目】国家自然科学基金(62071328）；天津市技术创新引导专项（21YDTPJC00540, 21YDTPJC00550） 【作者简介】李红利，博士，副教授，研究方向：生物信息检测与处理，E-mail: lihongliln@163.com
更新日期/Last Update: 2022-01-17