Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Charge based boundary element method with residual driven adaptive mesh refinement for high resolution electrical stimulation modeling.

Scientific reports·2026
Same author

Computer Vision for Monitoring Wild Bees and Wasps: A Structured Literature Review.

Ecology and evolution·2026
Same author

Influence of frequency and pulse train duration on respiratory responses during transcutaneous phrenic nerve stimulation in humans.

Journal of neural engineering·2026
Same author

Erratum: Multivariate assessment of the central-cardiorespiratory network structure in neuropathological disease (2018<i>Physiol. Meas</i>.<b>39</b>074004).

Physiological measurement·2026
Same author

An ultra-sensitive multi-channel MEG system for the non-invasive single-trial detection of cortical population spikes.

Scientific reports·2026
Same author

Muscle anisotropy influences the phrenic nerve activation threshold in non-invasive electrical stimulation.

Medical & biological engineering & computing·2026

Related Experiment Video

Updated: Aug 7, 2025

Author Spotlight: Advancing the Study of Brain-Heart Interplay with a Comprehensive EEGLAB Plugin for Multimodal Signal Analysis
08:22

Author Spotlight: Advancing the Study of Brain-Heart Interplay with a Comprehensive EEGLAB Plugin for Multimodal Signal Analysis

Published on: April 26, 2024

2.0K

A systematic comparison of deep learning methods for EEG time series analysis.

Dominik Walther1, Johannes Viehweg1, Jens Haueisen2

  • 1Data-Intensive Systems and Visualization Group (dAI.SY), Technische Universität Ilmenau, Ilmenau, Germany.

Frontiers in Neuroinformatics
|March 13, 2023
PubMed
Summary

Deep learning models like recurrent neural networks (RNNs) and feed-forward networks (FFNs) are compared for analyzing electroencephalography (EEG) time series. Temporal convolutional networks (TCNs) show superior performance on complex tasks, while LSTMs with attention excel on simpler ones.

Keywords:
attentionfeed forward neural networksrecurrent neural networkstime series analysistransformer networks

More Related Videos

Applications of EEG Neuroimaging Data: Event-related Potentials, Spectral Power, and Multiscale Entropy
11:15

Applications of EEG Neuroimaging Data: Event-related Potentials, Spectral Power, and Multiscale Entropy

Published on: June 27, 2013

33.8K
Analyzing Neural Activity and Connectivity Using Intracranial EEG Data with SPM Software
06:50

Analyzing Neural Activity and Connectivity Using Intracranial EEG Data with SPM Software

Published on: October 30, 2018

9.5K

Related Experiment Videos

Last Updated: Aug 7, 2025

Author Spotlight: Advancing the Study of Brain-Heart Interplay with a Comprehensive EEGLAB Plugin for Multimodal Signal Analysis
08:22

Author Spotlight: Advancing the Study of Brain-Heart Interplay with a Comprehensive EEGLAB Plugin for Multimodal Signal Analysis

Published on: April 26, 2024

2.0K
Applications of EEG Neuroimaging Data: Event-related Potentials, Spectral Power, and Multiscale Entropy
11:15

Applications of EEG Neuroimaging Data: Event-related Potentials, Spectral Power, and Multiscale Entropy

Published on: June 27, 2013

33.8K
Analyzing Neural Activity and Connectivity Using Intracranial EEG Data with SPM Software
06:50

Analyzing Neural Activity and Connectivity Using Intracranial EEG Data with SPM Software

Published on: October 30, 2018

9.5K

Area of Science:

  • Neuroscience
  • Machine Learning
  • Signal Processing

Background:

  • Analyzing electroencephalography (EEG) and magnetoencephalography (MEG) time series data is complex due to noise, high dimensionality, and patient-specific variations.
  • Deep learning methods, particularly recurrent neural networks (RNNs), show promise for time series analysis but can be computationally expensive and difficult to train.
  • Feed-forward networks (FFNs) often require handcrafted features, limiting their applicability to raw data analysis.

Purpose of the Study:

  • To systematically compare the performance of various deep learning architectures, including RNNs and FFNs, on EEG time series data.
  • To provide researchers with an updated guideline for selecting appropriate models for automated EEG analysis.
  • To establish a standardized experimental setup for fairer comparisons of different methodologies.

Main Methods:

  • A systematic comparison of Recurrent Neural Network (RNN) and Feed-Forward Network (FFN) topologies was conducted.
  • Advanced architectural concepts, including attention mechanisms and Temporal Convolutional Networks (TCNs), were evaluated.
  • All models were tested on multiple datasets using the same data preprocessing pipeline to ensure consistent experimental conditions.

Main Results:

  • A Long Short-Term Memory (LSTM) architecture with an attention mechanism achieved the best results on less complex tasks.
  • Temporal Convolutional Network (TCN) outperformed all recurrent architectures on the most complex dataset, showing an 8.61% accuracy improvement.
  • The attention mechanism significantly enhanced the classification performance of RNNs, and Extreme Learning Machines (ELMs) offered comparable results for simpler tasks.

Conclusions:

  • TCNs are highly effective for complex EEG analysis, while LSTMs with attention are suitable for less demanding tasks.
  • Attention mechanisms are crucial for improving RNN performance in EEG classification.
  • For lightweight and online learning applications, ELMs present a viable alternative for less complex EEG analysis tasks.