Search research articles

关于 JoVE

概览领导团队博客 JoVE 帮助中心

作者

出版流程编辑委员会范围与政策同行评审常见问题投稿

图书馆员

用户评价订阅访问资源图书馆顾问委员会常见问题

研究

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments 存档

教育

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual 教师资源中心教师网站

使用条款与条件

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序

Same author

Effect of fine structure of xanthan on its interaction with mucin: Toward designing of dysphagia diets.

Food chemistry·2026

Same author

Preparation and regulation of starch nanoparticles based on amino starch.

International journal of biological macromolecules·2026

Same author

Tailoring 3D-printed soy protein gels for dysphagia diets: Influence of xanthan gum's substituents and underlying mechanism.

International journal of biological macromolecules·2026

Same author

Radio frequency blanching facilitates enhanced release of bioactive phytochemicals and boosts antioxidant capacity in apple juice from Comminuted solid-liquid mixtures.

Food chemistry: X·2026

Same author

Robust Human-to-Robot Handover System Under Adverse Lighting.

Biomimetics (Basel, Switzerland)·2026

Same author

Ultra-Fast Infiltration Behavior of Vacuum Freeze-Dried Apples.

Small (Weinheim an der Bergstrasse, Germany)·2026

Search research articles

相关实验视频

Updated: Mar 15, 2026

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Published on: July 26, 2013

超学习增强多源域适应零校准电机图像EEG解码

Minmin Miao¹, Wenliang Fu¹, Hong Zeng²

¹School of Information Engineering, Huzhou University, Huzhou 313000, China.

Journal of neuroscience methods

|March 13, 2026

概括

这项研究引入了一种新的框架,用于无校准的运动图像大脑计算机接口 (MI-BCI) 解码. 超学习增强多源域适应 (MLEMSDA) 方法提高了中风神经康复应用的准确性.

关键词:

大脑计算机接口在EEG分类中,EEA的分类.超级学习 (Meta learning) 是一种超级学习.运动图像中的运动图像.转移学习转移学习

更多相关视频

Author Spotlight: Enhancing Neurorehabilitation Through EEG, Motor Imagery, and Virtual Reality

Author Spotlight: Enhancing Neurorehabilitation Through EEG, Motor Imagery, and Virtual Reality

Published on: May 10, 2024

Author Spotlight: Using Motor Imagery Brain-Computer Interface to Improve Motor and Cognitive Function in Stroke Patients

Author Spotlight: Using Motor Imagery Brain-Computer Interface to Improve Motor and Cognitive Function in Stroke Patients

Published on: September 1, 2023

相关实验视频

Last Updated: Mar 15, 2026

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Published on: July 26, 2013

Author Spotlight: Enhancing Neurorehabilitation Through EEG, Motor Imagery, and Virtual Reality

Author Spotlight: Enhancing Neurorehabilitation Through EEG, Motor Imagery, and Virtual Reality

Published on: May 10, 2024

Author Spotlight: Using Motor Imagery Brain-Computer Interface to Improve Motor and Cognitive Function in Stroke Patients

Author Spotlight: Using Motor Imagery Brain-Computer Interface to Improve Motor and Cognitive Function in Stroke Patients

Published on: September 1, 2023

科学领域:

神经科学是一个神经科学.
生物医学工程生物医学工程
机器学习机器学习

背景情况:

基于运动图像 (MI) 的脑计算机接口 (BCI) 显示了中风神经康复的潜力.
目前的MI-BCI系统面临的挑战是:主体间的变化,有限的训练数据和漫长的校准周期.

研究的目的:

开发一个新的框架,用于无校准MI-EEG解码.
解决现有的MI-BCI系统的局限性,包括学科间的变性和需要广泛的培训数据.

主要方法:

提出了一个Meta-Learning增强多源域适应 (MLEMSDA) 框架,统一跨任务,跨数据集和跨主题域适应.
基于梯度的元学习被用于无校准解码,利用对公共数据集的预训练和对目标数据集的元学习微调.
该框架在未见的受试者身上进行了测试,使用一个"留下一个"交叉验证方法.

主要成果:

在多个MI-EEG数据集上,MLEMSDA框架实现了高分类准确度:77.87% (在CBCIC上的DeepConvNet),75.54% (在自己的数据集上的EEGNet) 和72.72% (在BCI竞争IV数据集2b上的ShallowConvNet).
与竞争方法相比,拟议的方法在零校准场景中显示出更高的分类准确性.
验证是在公开和收集的MI-EEG数据集上进行的,包括中风患者.

结论:

在MLEMSDA框架有效地实现了无校准MI-EEG解码.
该方法显示出强大的通用性和有效性,为在神经康复中更实用的MI-BCI应用铺平了道路.