Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
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 JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

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

NeuroStat: An Open-Source EEG Connectivity Platform for Randomised Controlled Trials.

Sensors (Basel, Switzerland)·2026
Same author

Hydrodynamic Contrast Recanalization for Crossing Challenging Culprit Occlusions in Acute Myocardial Infarction.

JACC. Case reports·2026
Same author

Nonlinear EEG Complexity as a Marker of Maladaptive Brain Plasticity in Substance Use Disorders: A Multi-Group Machine Learning Classification Study.

Brain sciences·2026
Same author

Quantitative assessment of thoracolumbosacral orthosis efficacy on the temporal features of gait using inertial measurement unit sensors.

Scientific reports·2026
Same author

Nature-based solutions and indigenous knowledge for sustainable biological control.

Current opinion in insect science·2026
Same author

Oracle Upper Bounds on Clean-EEG Recoverability from Single-Channel Decompositions Under EOG/EMG Contamination.

Sensors (Basel, Switzerland)·2026

相关实验视频

Updated: Jul 5, 2025

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
08:15

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Published on: March 28, 2025

489

优化卷积神经网络的性能,以提高使用sEMG的手势识别.

Hassan Ashraf1, Asim Waris2, Syed Omer Gilani3

  • 1Laboratory of Movement Analysis (LAM-Motion Lab), University of Liège, Liège, Belgium.

Scientific reports
|January 24, 2024
PubMed
概括

这项研究使用贝叶斯优化和重叠数据分割技术优化了用于肌电控制的深度神经网络. 新方法显著降低了实时假肢控制应用的分类错误率.

更多相关视频

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

3.8K
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

542

相关实验视频

Last Updated: Jul 5, 2025

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
08:15

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Published on: March 28, 2025

489
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

3.8K
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

542

科学领域:

  • 生物医学工程 生物医学工程
  • 机器学习 机器学习
  • 信号处理 信号处理

背景情况:

  • 深度神经网络 (DNN) 对肌电控制 (MEC) 是有前途的,但由于优化而面临实时应用延迟.
  • 优化DNN超参数对于提高MEC系统性能和减少延迟至关重要.

研究的目的:

  • 研究基于卷积神经网络 (CNN) 的MEC系统的最佳配置.
  • 为基于DNN的MEC提出一个有效的数据细分技术和一套概括的超参数.

主要方法:

  • 与不同细分和重叠大小的分离和重叠数据细分策略进行了比较.
  • 采用贝叶斯优化来抽象和解决DNN超参数优化问题.
  • 收集了20名健康个体的表面电肌图 (sEMG) 数据,他们进行了10次抓取动作.

主要成果:

  • 叠加细分技术,最佳细分大小为200ms和80%的重叠,显著优于不连接的细分 (p <0.05).
  • 贝叶斯优化实现了0.08 ± 0.03的平均分类错误率 (CER),超过了手动,网格和随机搜索方法.
  • 在所有受试者中测试时,具有最佳超参数的通用CNN架构产生了0.09 ± 0.03的整体CER.

结论:

  • 在MEC应用中,重叠细分技术优越.
  • 贝叶斯优化为MEC系统中调整CNN超参数提供了一种有效的策略.
  • 拟议的通用CNN架构和优化的超参数为假肢控制和人机接口提供了实际的改进.