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

Multisensory integration in peripersonal space indexes consciousness states in sleep and disorders of consciousness.

Cell reports. Medicine·2026

Same author

Cortical alpha changes during visuospatial attention: a deep learning-enriched EEG analysis.

Cerebral cortex (New York, N.Y. : 1991)·2026

Same author

Cough acoustic analysis using artificial intelligence for COVID-19 detection: A comparative study of patient cohorts from Lima, Peru and Montreal, Canada.

Annals of epidemiology·2026

Same author

Inferring Arm Movement Direction from EEG Signals Using Explainable Deep Learning.

Sensors (Basel, Switzerland)·2026

Same author

On the virtues and limitations of Granger-causal brain connectivity estimate: Critical analysis using neural mass models.

Network neuroscience (Cambridge, Mass.)·2026

Same author

Revealing EEG signatures of intervention in disorder of consciousness using artificial intelligence: methodology and feasibility.

Computer methods and programs in biomedicine·2025

Search research articles

相关实验视频

Updated: Jun 7, 2025

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

用神经网络进行可靠的EEG解码的协议.

Davide Borra¹, Elisa Magosso¹, Mirco Ravanelli²

¹Department of Electrical, Electronic and Information Engineering "Guglielmo Marconi" (DEI), University of Bologna, Cesena, Forlì-Cesena, Italy.

Neural networks : the official journal of the International Neural Network Society

|November 16, 2024

概括

我们开发了一种用于脑电图 (EEG) 解码的综合协议,可以优化整个管道的超参数. 这种方法通过最小化性能波动和提高准确性来确保可靠和值得信赖的EEG解码.

关键词:

大脑计算机接口卷积神经网络是一种卷积神经网络.深度学习是一种深度学习.电脑电图 (电脑电图) 是一种脑电图.超参数搜索搜索可以通过超参数搜索进行.一次试验的EEG解码.

更多相关视频

Network Analysis of Foramen Ovale Electrode Recordings in Drug-resistant Temporal Lobe Epilepsy Patients

Network Analysis of Foramen Ovale Electrode Recordings in Drug-resistant Temporal Lobe Epilepsy Patients

Published on: December 18, 2016

A Novel Experimental and Analytical Approach to the Multimodal Neural Decoding of Intent During Social Interaction in Freely-behaving Human Infants

A Novel Experimental and Analytical Approach to the Multimodal Neural Decoding of Intent During Social Interaction in Freely-behaving Human Infants

Published on: October 4, 2015

相关实验视频

Last Updated: Jun 7, 2025

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

Network Analysis of Foramen Ovale Electrode Recordings in Drug-resistant Temporal Lobe Epilepsy Patients

Network Analysis of Foramen Ovale Electrode Recordings in Drug-resistant Temporal Lobe Epilepsy Patients

Published on: December 18, 2016

A Novel Experimental and Analytical Approach to the Multimodal Neural Decoding of Intent During Social Interaction in Freely-behaving Human Infants

A Novel Experimental and Analytical Approach to the Multimodal Neural Decoding of Intent During Social Interaction in Freely-behaving Human Infants

Published on: October 4, 2015

科学领域:

神经科学是一个神经科学.
机器学习机器学习
信号处理信号处理

背景情况:

深度学习模型在电脑电图 (EEG) 解码方面实现了最先进的性能.
现有的EEG解码管道存在许多超参数和对随机初始化的敏感性,影响了可靠性.
自动超参数搜索是有限的,很少覆盖整个解码管道.

研究的目的:

设计和验证一个全面的EEG解码协议,以优化整个管道中的超参数.
通过多种子初始化,确保可靠的性能估计.
建立一个值得信赖和可靠的标准方法来解码EEG.

主要方法:

开发了一个全面的超参数搜索协议,涵盖数据预处理,网络架构,培训和增强.
集成的多种子初始化可用于可靠的性能估计.
在9个不同的EEG数据集 (运动图像,P300,SSVEP) 上验证了协议,其中有204名参与者,探索搜索策略,参与者子集和算法类型.

主要成果:

最优的协议涉及使用知情算法进行双步超参数搜索和10次随机初始化进行最终培训和评估.
使用3-5名参与者进行超参数搜索,实现了性能和计算成本之间的最佳平衡.
拟议的协议在各种数据集和深度学习模型中始终超过基线最先进的管道.

结论:

开发的综合协议提高了EEG解码的可靠性和可信度.
该协议为神经科学家提供了一种标准化的方法来优化EEG解码管道.
该方法解决了当前基于深度学习的EEG解码解决方案的关键局限性.