• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利：CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法，用中文像聊天一样搜索，搜遍4000万医学文献。AI智能推荐，让科研检索更轻松。

立即免费搜索

文件翻译

保留排版，准确专业，支持PDF/Word/PPT等文件格式，支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述，25分钟生成高质量综述，智能提取关键信息，辅助科研写作。

立即免费体验

用于情感识别中脑电通道约简的堆叠自动编码器特征提取

Feature Extraction With Stacked Autoencoders for EEG Channel Reduction in Emotion Recognition.

作者信息

Vafaei Elnaz, Nowshiravan Rahatabad Fereidoun, Setarehdan Seyed Kamaledin, Azadfallah Parviz

机构信息

Department of Biomedical Engineering, Faculty of Medical Sciences and Technologies, Science and Research Branch, Islamic Azad University, Tehran, Iran.

School of Electrical and Computer Engineering, University of Tehran, Tehran, Iran.

出版信息

Basic Clin Neurosci. 2024 May-Jun;15(3):393-402. doi: 10.32598/bcn.2023.5138.2. Epub 2024 May 1.

DOI:10.32598/bcn.2023.5138.2
PMID:39403356
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC11470895/
Abstract

INTRODUCTION

Emotion recognition by electroencephalogram (EEG) signals is one of the complex methods because the extraction and recognition of the features hidden in the signal are sophisticated and require a significant number of EEG channels. Presenting a method for feature analysis and an algorithm for reducing the number of EEG channels fulfills the need for research in this field.

METHODS

Accordingly, this study investigates the possibility of utilizing deep learning to reduce the number of channels while maintaining the quality of the EEG signal. A stacked autoencoder network extracts optimal features for emotion classification in valence and arousal dimensions. Autoencoder networks can extract complex features to provide linear and non- linear features which are a good representative of the signal.

RESULTS

The accuracy of a conventional emotion recognition classifier (support vector machine) using features extracted from SAEs was obtained at 75.7% for valence and 74.4% for arousal dimensions, respectively.

CONCLUSION

Further analysis also illustrates that valence dimension detection with reduced EEG channels has a different composition of EEG channels compared to the arousal dimension. In addition, the number of channels is reduced from 32 to 12, which is an excellent development for designing a small-size EEG device by applying these optimal features.

摘要

引言

通过脑电图（EEG）信号进行情绪识别是一种复杂的方法，因为从信号中提取和识别隐藏特征非常复杂，并且需要大量的EEG通道。提出一种特征分析方法和一种减少EEG通道数量的算法满足了该领域的研究需求。

方法

因此，本研究探讨了利用深度学习在保持EEG信号质量的同时减少通道数量的可能性。一个堆叠自动编码器网络提取用于效价和唤醒维度情绪分类的最优特征。自动编码器网络可以提取复杂特征以提供线性和非线性特征，这些特征是信号的良好代表。

结果

使用从堆叠自动编码器提取的特征的传统情绪识别分类器（支持向量机）的准确率在效价维度为75.7%，在唤醒维度为74.4%。

结论

进一步分析还表明，与唤醒维度相比，减少EEG通道的效价维度检测具有不同的EEG通道组成。此外，通道数量从32个减少到12个，这对于通过应用这些最优特征设计小型EEG设备来说是一个出色的进展。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/41a2/11470895/1311aa2e6218/BCN-15-393-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/41a2/11470895/874fcd815030/BCN-15-393-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/41a2/11470895/6964cfb108ef/BCN-15-393-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/41a2/11470895/026cfa6705e4/BCN-15-393-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/41a2/11470895/1311aa2e6218/BCN-15-393-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/41a2/11470895/874fcd815030/BCN-15-393-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/41a2/11470895/6964cfb108ef/BCN-15-393-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/41a2/11470895/026cfa6705e4/BCN-15-393-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/41a2/11470895/1311aa2e6218/BCN-15-393-g004.jpg

相似文献

1
Feature Extraction With Stacked Autoencoders for EEG Channel Reduction in Emotion Recognition.用于情感识别中脑电通道约简的堆叠自动编码器特征提取
Basic Clin Neurosci. 2024 May-Jun;15(3):393-402. doi: 10.32598/bcn.2023.5138.2. Epub 2024 May 1.
2
An Ensemble Learning Method for Emotion Charting Using Multimodal Physiological Signals.基于多模态生理信号的情绪图表分析的集成学习方法。
Sensors (Basel). 2022 Dec 4;22(23):9480. doi: 10.3390/s22239480.
3
Electroencephalogram Emotion Recognition Based on 3D Feature Fusion and Convolutional Autoencoder.基于三维特征融合与卷积自动编码器的脑电图情感识别
Front Comput Neurosci. 2021 Oct 18;15:743426. doi: 10.3389/fncom.2021.743426. eCollection 2021.
4
Emotion recognition with reduced channels using CWT based EEG feature representation and a CNN classifier.基于连续小波变换的脑电图特征表示和卷积神经网络分类器的少通道情感识别
Biomed Phys Eng Express. 2024 Apr 30;10(4). doi: 10.1088/2057-1976/ad31f9.
5
Multi-channel EEG-based emotion recognition via a multi-level features guided capsule network.基于多通道 EEG 的多水平特征引导胶囊网络情绪识别。
Comput Biol Med. 2020 Aug;123:103927. doi: 10.1016/j.compbiomed.2020.103927. Epub 2020 Jul 22.
6
Extracting a Novel Emotional EEG Topographic Map Based on a Stacked Autoencoder Network.基于堆叠自编码器网络的新型情绪 EEG 地形图谱提取。
J Healthc Eng. 2023 Jan 19;2023:9223599. doi: 10.1155/2023/9223599. eCollection 2023.
7
Emotion recognition from multichannel EEG signals using K-nearest neighbor classification.使用K近邻分类法从多通道脑电图信号中进行情绪识别。
Technol Health Care. 2018;26(S1):509-519. doi: 10.3233/THC-174836.
8
Comprehensive Analysis of Feature Extraction Methods for Emotion Recognition from Multichannel EEG Recordings.多通道 EEG 记录的情感识别特征提取方法的综合分析。
Sensors (Basel). 2023 Jan 12;23(2):915. doi: 10.3390/s23020915.
9
EEG emotion recognition using reduced channel wavelet entropy and average wavelet coefficient features with normal Mutual Information method.基于约简通道小波熵和平均小波系数特征并采用正态互信息方法的脑电图情感识别
Annu Int Conf IEEE Eng Med Biol Soc. 2017 Jul;2017:463-466. doi: 10.1109/EMBC.2017.8036862.
10
EEG emotion recognition based on data-driven signal auto-segmentation and feature fusion.基于数据驱动信号自动分割与特征融合的脑电图情感识别
J Affect Disord. 2024 Sep 15;361:356-366. doi: 10.1016/j.jad.2024.06.042. Epub 2024 Jun 15.

引用本文的文献

1
Transformers in EEG Analysis: A Review of Architectures and Applications in Motor Imagery, Seizure, and Emotion Classification.脑电图分析中的变压器：运动想象、癫痫发作及情绪分类的架构与应用综述
Sensors (Basel). 2025 Feb 20;25(5):1293. doi: 10.3390/s25051293.

本文引用的文献

1
Electroencephalograph Emotion Classification Using a Novel Adaptive Ensemble Classifier Considering Personality Traits.基于考虑人格特质的新型自适应集成分类器的脑电图情感分类
Basic Clin Neurosci. 2023 Sep-Oct;14(5):687-700. doi: 10.32598/bcn.2022.3830.2. Epub 2023 Sep 1.
2
Automated EEG pathology detection based on different convolutional neural network models: Deep learning approach.基于不同卷积神经网络模型的自动 EEG 病理检测：深度学习方法。
Comput Biol Med. 2021 Jun;133:104434. doi: 10.1016/j.compbiomed.2021.104434. Epub 2021 Apr 25.
3
EEG emotion recognition using reduced channel wavelet entropy and average wavelet coefficient features with normal Mutual Information method.
基于约简通道小波熵和平均小波系数特征并采用正态互信息方法的脑电图情感识别
Annu Int Conf IEEE Eng Med Biol Soc. 2017 Jul;2017:463-466. doi: 10.1109/EMBC.2017.8036862.
4
Recognition of emotions using multimodal physiological signals and an ensemble deep learning model.利用多模态生理信号和集成深度学习模型识别情绪。
Comput Methods Programs Biomed. 2017 Mar;140:93-110. doi: 10.1016/j.cmpb.2016.12.005. Epub 2016 Dec 15.
5
Emotion Perception from Face, Voice, and Touch: Comparisons and Convergence.面部、声音和触觉的情绪感知：比较与融合
Trends Cogn Sci. 2017 Mar;21(3):216-228. doi: 10.1016/j.tics.2017.01.001. Epub 2017 Feb 4.
6
Unsupervised domain adaptation techniques based on auto-encoder for non-stationary EEG-based emotion recognition.基于自动编码器的非平稳 EEG 基情绪识别无监督领域自适应技术。
Comput Biol Med. 2016 Dec 1;79:205-214. doi: 10.1016/j.compbiomed.2016.10.019. Epub 2016 Oct 22.