[1]王安琪,于超,陈胤玮,等.基于LSTM-Transformer的脑电情感分析[J].中国医学物理学杂志,2024,41(12):1550-1557.[doi:DOI:10.3969/j.issn.1005-202X.2024.12.013]
　WANG Anqi,YU Chao,CHEN Yinwei,et al.EEG emotion analysis based on LSTM-Transformer[J].Chinese Journal of Medical Physics,2024,41(12):1550-1557.[doi:DOI:10.3969/j.issn.1005-202X.2024.12.013]
点击复制

基于LSTM-Transformer的脑电情感分析()
分享到：

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

卷:
41卷
期数:
2024年第12期
页码:
1550-1557
栏目:
医学信号处理与医学仪器
出版日期:
2024-12-17

文章信息/Info

Title:
EEG emotion analysis based on LSTM-Transformer
文章编号:
1005-202X（2024）12-1550-08
作者:
王安琪1; 于超1; 陈胤玮1; 郗群2
1.甘肃省中医药大学信息工程学院， 甘肃 兰州 730000； 2.兰州大学第二医院信息中心， 甘肃 兰州 730001
Author(s):
WANG Anqi1;  YU Chao1;  CHEN Yinwei1;  XI Qun2
1.School of Information Engineering, Gansu University of Chinese Medicine, Lanzhou 730000, China 2.Information Center, the Second Hospital of Lanzhou University, Lanzhou 730001, China
关键词:
脑电; 情绪识别; 深度学习; LSTM; Transformer
Keywords:
Keywords: electroencephalogram emotion recognition deep learning long short-term memory Transformer
分类号:
R318；TP391.7
DOI:
DOI:10.3969/j.issn.1005-202X.2024.12.013
文献标志码:
A
摘要:
针对传统情感识别方法在处理长期依赖关系时的不足，提出一种结合长短期记忆网络（LSTM）与Transformer模块的脑电情感分析模型，称为LTNet。该模型首先对数据进行预处理，然后将提取的特征输入至LTNet。LSTM模块和Transformer模块独立对输入的序列进行建模，分别从中提取出深层的局部特征和全局特征。通过采用加权融合策略来综合这些特征，最终利用Softmax函数对情绪进行四分类。实验结果显示，在DEAP数据集上进行的五折交叉验证中，LTNet的平均识别准确率达到96.56%，相比于传统机器学习算法和其他深度学习方法提高2.74%~21.31%。
Abstract:
Abstract: An electroencephalogram (EEG) emotion analysis model (LTNet) that combines long short-term memory (LSTM) and Transformer modules is proposed for addressing the shortcomings of traditional emotion recognition methods in dealing with long-term dependencies. After data preprocessing, the extracted features are input into LTNet. LSTM module and Transformer module model the input sequence independently, and from which deep local features and global features are extracted and then fused using a weighted fusion strategy. Finally, Softmax function is used to classify emotions into 4 categories. Experimental results show that LTNet has an average recognition accuracy of 96.56% in the 5-fold cross-validation on the DEAP dataset, which is 2.74%-21.31% higher than traditional machine learning algorithms and other deep learning methods.

相似文献/References:

[1]杨芳,李涛.基于能量特征和模糊熵的睡眠自动分期[J].中国医学物理学杂志,2017,34(9):968.[doi:DOI:10.3969/j.issn.1005-202X.2017.09.022]
　[J].Chinese Journal of Medical Physics,2017,34(12):968.[doi:DOI:10.3969/j.issn.1005-202X.2017.09.022]
[2]丁正敏,熊冬生,陈宇珂,等. 基于脑电样本熵和小波熵的麻醉深度监测[J].中国医学物理学杂志,2018,35(2):243.[doi:DOI:10.3969/j.issn.1005-202X.2018.02.024]
　DING Zhengmin,XIONG Dongsheng,CHEN Yuke,et al. Sample entropy and wavelet entropy of electroencephalogram for monitoring the depth of anesthesia[J].Chinese Journal of Medical Physics,2018,35(12):243.[doi:DOI:10.3969/j.issn.1005-202X.2018.02.024]
[3]欧朗,赵越,李晓欧.NIRS-EEG联合检测技术在脑功能分析中的研究进展[J].中国医学物理学杂志,2020,37(10):1323.[doi:DOI:10.3969/j.issn.1005-202X.2020.10.019]
　OU Lang,ZHAO Yue,LI Xiaoou.Advances of near infrared spectroscopy combined with electroencephalogram in brain function analysls[J].Chinese Journal of Medical Physics,2020,37(12):1323.[doi:DOI:10.3969/j.issn.1005-202X.2020.10.019]
[4]杨兰,随力,刘亮,等.不同视错觉刺激诱发的事件相关电位的脑电研究进展[J].中国医学物理学杂志,2021,38(5):613.[doi:DOI:10.3969/j.issn.1005-202X.2021.05.017]
　YANG Lan,SUI Li,LIU Liang,et al.Advances in research on event-related potentials evoked by different visual illusions[J].Chinese Journal of Medical Physics,2021,38(12):613.[doi:DOI:10.3969/j.issn.1005-202X.2021.05.017]
[5]李凌,张婷婷,谢珂.基于Unity3D的脑电图检测与分析虚拟仿真实验项目的设计和开发[J].中国医学物理学杂志,2021,38(5):631.[doi:DOI:10.3969/j.issn.1005-202X.2021.05.020]
　LI Ling,ZHANG Tingting,XIE Ke.Design and development of a Unity3D-based virtual simulation experiment platform for electroencephalogram detection and analysis[J].Chinese Journal of Medical Physics,2021,38(12):631.[doi:DOI:10.3969/j.issn.1005-202X.2021.05.020]
[6]陈枭宇,随力,李丽.脑电神经反馈训练在提高工作记忆中应用的研究与挑战[J].中国医学物理学杂志,2022,39(8):987.[doi:DOI:10.3969/j.issn.1005-202X.2022.08.012]
　CHEN Xiaoyu,SUI Li,LI Li.Advances and challenges of EEG neurofeedback training in improving working memory[J].Chinese Journal of Medical Physics,2022,39(12):987.[doi:DOI:10.3969/j.issn.1005-202X.2022.08.012]
[7]吴彦泽,王海玲,高宇飞.基于脑电的脑网络稳定模式情绪识别研究[J].中国医学物理学杂志,2023,40(6):727.[doi:DOI:10.3969/j.issn.1005-202X.2023.06.010]
　WU Yanze,WANG Hailing,GAO Yufei.Stable brain network pattern for EEG-based emotion recognition[J].Chinese Journal of Medical Physics,2023,40(12):727.[doi:DOI:10.3969/j.issn.1005-202X.2023.06.010]
[8]王冲,周静,李文豪,等.脑电在脑卒中运动功能评估中的应用研究进展[J].中国医学物理学杂志,2023,40(10):1309.[doi:DOI:10.3969/j.issn.1005-202X.2023.10.020]
　WANG Chong,ZHOU Jing,LI Wenhao,et al.Application of electroencephalography in assessment of motor function in stroke patients: a review[J].Chinese Journal of Medical Physics,2023,40(12):1309.[doi:DOI:10.3969/j.issn.1005-202X.2023.10.020]
[9]任志伟,马宇欣,薛婷,等.EEG在尼古丁成瘾领域中的应用研究[J].中国医学物理学杂志,2024,41(12):1537.[doi:DOI:10.3969/j.issn.1005-202X.2024.12.011]
　REN Zhiwei,MA Yuxin,XUE Ting,et al.Application of EEG in nicotine addiction research[J].Chinese Journal of Medical Physics,2024,41(12):1537.[doi:DOI:10.3969/j.issn.1005-202X.2024.12.011]

备注/Memo

备注/Memo:
【收稿日期】2024-07-08 【基金项目】甘肃省自然科学基金（20CX9JA145） 【作者简介】王安琪，硕士，研究方向：脑电情绪识别，E-mail:1101709211@qq.com
更新日期/Last Update: 2024-12-20