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

The Role of Indocyanine Green Videoangiography in Intramedullary Spinal Cord Tumor Surgery: Focus on Lesion Localization Accuracy.

Neurologia medico-chirurgica·2026
Same author

Structural Ordering and Polarizable Al<sup>3+</sup> Solvation Induced via Amphiphilic Zwitterion for Highly-Reversible Zn-Al Alloy Anodes.

Small (Weinheim an der Bergstrasse, Germany)·2026
Same author

Spatiotemporal Dynamics in Prespeech Semantic Category Decoding: An Intracranial EEG Study.

eNeuro·2026
Same author

Long-term radiological progression after resection of dysembryoplastic neuroepithelial tumors: patterns and prognostic factors.

Journal of neuro-oncology·2026
Same author

Figure-ground relationship of voices in musical structure modulates reciprocal frontotemporal connectivity.

Frontiers in neuroscience·2026
Same author

Prognostic factors for spinal cord cavernous hemangioma resection and optimal surgical timing for symptomatic patients.

Journal of neurosurgery. Spine·2026

相关实验视频

Updated: Jun 22, 2025

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces
10:51

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

Published on: March 10, 2011

13.7K

双模型转移学习以弥补大脑-计算机接口的个体变化.

Jun Su Kim1, HongJune Kim2, Chun Kee Chung3

  • 1Dept. of Brain and Cognitive Sciences, Seoul National University, Seoul, Republic of Korea; Clinical Research Institute, Konkuk University Medical Center Seoul, Republic of Korea.

Computer methods and programs in biomedicine
|June 29, 2024
PubMed
概括

这项研究引入了一种用于脑计算机接口 (BCI) 的新型转移学习方法,该方法通过结合个人和群体深度神经网络模型来提高解码性能,有效地解决特定主体的变化.

关键词:
大脑与计算机接口 (BCI)在深深的深处,深深的深处.不同质的数据 不同质的数据个体变异性是个人的变异性.神经网络 (DNN) 是一个神经网络.来源估计来源估计转移学习转移学习

更多相关视频

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding
11:25

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

Published on: July 26, 2013

43.3K
Assessment and Communication for People with Disorders of Consciousness
07:37

Assessment and Communication for People with Disorders of Consciousness

Published on: August 1, 2017

9.0K

相关实验视频

Last Updated: Jun 22, 2025

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces
10:51

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

Published on: March 10, 2011

13.7K
Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding
11:25

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

Published on: July 26, 2013

43.3K
Assessment and Communication for People with Disorders of Consciousness
07:37

Assessment and Communication for People with Disorders of Consciousness

Published on: August 1, 2017

9.0K

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:

  • 大脑计算机接口 (BCI) 技术的进步越来越多地利用深度神经网络 (DNN) 来完成复杂的解码任务,例如任意运动回归.
  • 在单个数据上训练的BCI模型通常由于大量DNN参数和广泛的数据集要求而受到有限的性能和糟糕的概括性影响.
  • 群组数据可能不会产生足够的解码性能,因为神经信号在个人之间和随着时间的推移内在的变化.

研究的目的:

  • 开发一种转移学习方法,有效地适应皮层区域的特定学科变异性,以提高BCI性能.
  • 创建一个强大的BCI解码模型,结合个人和团体数据分析的优势.

主要方法:

  • 训练单独的运动解码模型在个人和聚合组数据上.
  • 从单个模型生成突出地图,以识别不同主题的输入贡献差异.
  • 使用修改的知识蒸框架,加权通用适用性和个人微调的组合个体和组模型.

主要成果:

  • 拟议的组合模型表现出优越的解码性能 (平均r = 0.75),与个人 (r = 0.70) 和组模型 (r = 0.40) 相比,在手臂伸展任务中表现出优异的解码性能.
  • 在个别模型最初显示低解码精度的情况下 (例如,从r = 0.50到r = 0.61),观察到显著的性能改进.
  • 该方法有效地封装并适应个人神经信号的变化.

结论:

  • 开发的转移学习方法显示出在创建强大且广泛适用的脑计算机接口方面有很大的潜力.
  • 该方法成功地将个人BCI数据概括起来,提高解码性能,克服传统个人或组模型的局限性.