• 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

基于深度学习方法的想象元音脑电信号识别。

Recognition of EEG Signals from Imagined Vowels Using Deep Learning Methods.

机构信息

Departamento de Tecnología, Universidad Pedagógica Nacional, Bogotá 111321, Colombia.

Department of Electrical and Electronics Engineering, School of Engineering, Universidad Nacional de Colombia, Bogotá 111321, Colombia.

出版信息

Sensors (Basel). 2021 Sep 29;21(19):6503. doi: 10.3390/s21196503.

DOI:10.3390/s21196503
PMID:34640824
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC8512781/
Abstract

The use of imagined speech with electroencephalographic (EEG) signals is a promising field of brain-computer interfaces (BCI) that seeks communication between areas of the cerebral cortex related to language and devices or machines. However, the complexity of this brain process makes the analysis and classification of this type of signals a relevant topic of research. The goals of this study were: to develop a new algorithm based on Deep Learning (DL), referred to as CNNeeg1-1, to recognize EEG signals in imagined vowel tasks; to create an imagined speech database with 50 subjects specialized in imagined vowels from the Spanish language (/a/,/e/,/i/,/o/,/u/); and to contrast the performance of the CNNeeg1-1 algorithm with the DL Shallow CNN and EEGNet benchmark algorithms using an open access database (BD1) and the newly developed database (BD2). In this study, a mixed variance analysis of variance was conducted to assess the intra-subject and inter-subject training of the proposed algorithms. The results show that for intra-subject training analysis, the best performance among the Shallow CNN, EEGNet, and CNNeeg1-1 methods in classifying imagined vowels (/a/,/e/,/i/,/o/,/u/) was exhibited by CNNeeg1-1, with an accuracy of 65.62% for BD1 database and 85.66% for BD2 database.

摘要

使用脑电图 (EEG) 信号进行想象性言语是脑机接口 (BCI) 的一个有前途的领域，它旨在实现与语言相关的大脑皮层区域和设备或机器之间的交流。然而，这种大脑过程的复杂性使得对这种类型的信号的分析和分类成为一个相关的研究课题。本研究的目的是：开发一种基于深度学习 (DL) 的新算法，称为 CNNeeg1-1，用于识别想象性元音任务中的 EEG 信号；创建一个包含 50 名专门从事西班牙语想象元音的想象性语音数据库 (/a/、/e/、/i/、/o/、/u/)；并使用开放访问数据库 (BD1) 和新开发的数据库 (BD2) ，对比 CNNeeg1-1 算法与 DL Shallow CNN 和 EEGNet 基准算法的性能。在这项研究中，进行了混合方差分析，以评估所提出算法的个体内和个体间训练。结果表明，对于个体内训练分析，在对想象性元音 (/a/、/e/、/i/、/o/、/u/)进行分类时，Shallow CNN、EEGNet 和 CNNeeg1-1 方法中表现最好的是 CNNeeg1-1，BD1 数据库的准确率为 65.62%，BD2 数据库的准确率为 85.66%。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/3c19c5b9bfda/sensors-21-06503-g015.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/804b9fef3c79/sensors-21-06503-g0A1.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/3714725c66db/sensors-21-06503-g0A2.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/8f0ac48aba40/sensors-21-06503-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/8839917ff818/sensors-21-06503-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/8b972f5203bd/sensors-21-06503-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/1512515be7cb/sensors-21-06503-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/7add93abe52d/sensors-21-06503-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/6bcf44f7c4d1/sensors-21-06503-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/e56bb5405321/sensors-21-06503-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/2a6a9dc3067b/sensors-21-06503-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/2c5ea84943ce/sensors-21-06503-g009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/5d4c508ea45a/sensors-21-06503-g010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/1a9d283e5f41/sensors-21-06503-g011.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/bba7b350a0bc/sensors-21-06503-g012.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/1d7f3b89d454/sensors-21-06503-g013.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/4f5951cd265d/sensors-21-06503-g014.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/3c19c5b9bfda/sensors-21-06503-g015.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/804b9fef3c79/sensors-21-06503-g0A1.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/3714725c66db/sensors-21-06503-g0A2.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/8f0ac48aba40/sensors-21-06503-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/8839917ff818/sensors-21-06503-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/8b972f5203bd/sensors-21-06503-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/1512515be7cb/sensors-21-06503-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/7add93abe52d/sensors-21-06503-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/6bcf44f7c4d1/sensors-21-06503-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/e56bb5405321/sensors-21-06503-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/2a6a9dc3067b/sensors-21-06503-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/2c5ea84943ce/sensors-21-06503-g009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/5d4c508ea45a/sensors-21-06503-g010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/1a9d283e5f41/sensors-21-06503-g011.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/bba7b350a0bc/sensors-21-06503-g012.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/1d7f3b89d454/sensors-21-06503-g013.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/4f5951cd265d/sensors-21-06503-g014.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/88e1/8512781/3c19c5b9bfda/sensors-21-06503-g015.jpg

相似文献

1
Recognition of EEG Signals from Imagined Vowels Using Deep Learning Methods.基于深度学习方法的想象元音脑电信号识别。
Sensors (Basel). 2021 Sep 29;21(19):6503. doi: 10.3390/s21196503.
2
Evaluation of Hyperparameter Optimization in Machine and Deep Learning Methods for Decoding Imagined Speech EEG.机器和深度学习方法在解码想象语音 EEG 中的超参数优化评估。
Sensors (Basel). 2020 Aug 17;20(16):4629. doi: 10.3390/s20164629.
3
Resting state EEG assisted imagined vowel phonemes recognition by native and non-native speakers using brain connectivity measures.基于脑连接测量的静息态 EEG 辅助母语和非母语者想象元音音位识别。
Phys Eng Sci Med. 2024 Sep;47(3):939-954. doi: 10.1007/s13246-024-01417-w. Epub 2024 Apr 22.
4
EEG-based classification of imagined digits using a recurrent neural network.基于脑电图的循环神经网络对想象数字的分类
J Neural Eng. 2023 Apr 28;20(2). doi: 10.1088/1741-2552/acc976.
5
Enhancing Cross-Subject Motor Imagery Classification in EEG-Based Brain-Computer Interfaces by Using Multi-Branch CNN.基于多分支卷积神经网络的 EEG 脑-机接口中跨被试运动想象分类的增强。
Sensors (Basel). 2023 Sep 15;23(18):7908. doi: 10.3390/s23187908.
6
EEGNet: a compact convolutional neural network for EEG-based brain-computer interfaces.EEGNet：一种基于 EEG 的脑机接口用的紧凑卷积神经网络。
J Neural Eng. 2018 Oct;15(5):056013. doi: 10.1088/1741-2552/aace8c. Epub 2018 Jun 22.
7
Imagined character recognition through EEG signals using deep convolutional neural network.基于深度卷积神经网络的脑电信号想象字符识别。
Med Biol Eng Comput. 2021 May;59(5):1167-1183. doi: 10.1007/s11517-021-02368-0. Epub 2021 May 4.
8
A Bimodal Deep Learning Architecture for EEG-fNIRS Decoding of Overt and Imagined Speech.一种用于 EEG-fNIRS 解码言语出声和想象的双模深度学习架构。
IEEE Trans Biomed Eng. 2022 Jun;69(6):1983-1994. doi: 10.1109/TBME.2021.3132861. Epub 2022 May 19.
9
Inter-subject transfer learning with an end-to-end deep convolutional neural network for EEG-based BCI.基于端到端深度卷积神经网络的 EEG 脑机接口的跨被试迁移学习。
J Neural Eng. 2019 Apr;16(2):026007. doi: 10.1088/1741-2552/aaf3f6. Epub 2018 Nov 26.
10
Vowel Imagery Decoding toward Silent Speech BCI Using Extreme Learning Machine with Electroencephalogram.基于脑电的极端学习机的元音想象解码在无声语音脑机接口中的应用。
Biomed Res Int. 2016;2016:2618265. doi: 10.1155/2016/2618265. Epub 2016 Dec 19.

引用本文的文献

1
From pronounced to imagined: improving speech decoding with multi-condition EEG data.从语音发音到语音想象：利用多条件脑电图数据改善语音解码
Front Neuroinform. 2025 Jun 27;19:1583428. doi: 10.3389/fninf.2025.1583428. eCollection 2025.
2
Lobish: Symbolic Language for Interpreting Electroencephalogram Signals in Language Detection Using Channel-Based Transformation and Pattern.Lobish：用于在基于通道变换和模式的语言检测中解释脑电图信号的符号语言。
Diagnostics (Basel). 2024 Sep 8;14(17):1987. doi: 10.3390/diagnostics14171987.
3
Decoding imagined speech with delay differential analysis.

本文引用的文献

1
Sparse Representation-Based Denoising for High-Resolution Brain Activation and Functional Connectivity Modeling: A Task fMRI Study.基于稀疏表示的去噪用于高分辨率脑激活和功能连接建模：一项任务功能磁共振成像研究
IEEE Access. 2020;8:36728-36740. doi: 10.1109/access.2020.2971261. Epub 2020 Feb 3.
2
Model-Based Deep Learning PET Image Reconstruction Using Forward-Backward Splitting Expectation-Maximization.基于模型的深度学习PET图像重建：使用前向-后向分裂期望最大化算法
IEEE Trans Radiat Plasma Med Sci. 2020 Jun 23;5(1):54-64. doi: 10.1109/TRPMS.2020.3004408.
3
Learning Subject-Generalized Topographical EEG Embeddings Using Deep Variational Autoencoders and Domain-Adversarial Regularization.
利用延迟微分分析解码想象中的言语。
Front Hum Neurosci. 2024 May 17;18:1398065. doi: 10.3389/fnhum.2024.1398065. eCollection 2024.
4
Advanced Modeling and Signal Processing Methods in Brain-Computer Interfaces Based on a Vector of Cyclic Rhythmically Connected Random Processes.基于循环节律连接随机过程向量的脑机接口中的高级建模与信号处理方法。
Sensors (Basel). 2023 Jan 9;23(2):760. doi: 10.3390/s23020760.
5
A Novel Stress State Assessment Method for College Students Based on EEG.基于脑电的大学生新型应激状态评估方法。
Comput Intell Neurosci. 2022 Jun 7;2022:4565968. doi: 10.1155/2022/4565968. eCollection 2022.
使用深度变分自动编码器和领域对抗正则化学习主题泛化拓扑脑电图嵌入。
Sensors (Basel). 2021 Mar 4;21(5):1792. doi: 10.3390/s21051792.
4
A Comprehensive Review on Critical Issues and Possible Solutions of Motor Imagery Based Electroencephalography Brain-Computer Interface.基于运动想象的脑-机接口关键问题及可能解决方案的综合评述
Sensors (Basel). 2021 Mar 20;21(6):2173. doi: 10.3390/s21062173.
5
Biosignal Sensors and Deep Learning-Based Speech Recognition: A Review.生物信号传感器与基于深度学习的语音识别：综述。
Sensors (Basel). 2021 Feb 17;21(4):1399. doi: 10.3390/s21041399.
6
The MindGomoku: An Online P300 BCI Game Based on Bayesian Deep Learning.思维五子棋：一款基于贝叶斯深度学习的在线P300脑机接口游戏。
Sensors (Basel). 2021 Feb 25;21(5):1613. doi: 10.3390/s21051613.
7
An Open Dataset for Wearable SSVEP-Based Brain-Computer Interfaces.基于可穿戴 SSVEP 的脑机接口的公开数据集。
Sensors (Basel). 2021 Feb 10;21(4):1256. doi: 10.3390/s21041256.
8
Precision Assessment of COVID-19 Phenotypes Using Large-Scale Clinic Visit Audio Recordings: Harnessing the Power of Patient Voice.利用大规模临床就诊音频记录精准评估 COVID-19 表型：挖掘患者声音的力量。
J Med Internet Res. 2021 Feb 19;23(2):e20545. doi: 10.2196/20545.
9
Developing a Motor Imagery-Based Real-Time Asynchronous Hybrid BCI Controller for a Lower-Limb Exoskeleton.开发一种用于下肢外骨骼的基于运动想象的实时异步混合脑机接口控制器。
Sensors (Basel). 2020 Dec 19;20(24):7309. doi: 10.3390/s20247309.
10
Single-Option P300-BCI Performance Is Affected by Visual Stimulation Conditions.单一选择 P300-BCI 性能受视觉刺激条件影响。
Sensors (Basel). 2020 Dec 16;20(24):7198. doi: 10.3390/s20247198.