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

Complex wavelet-based Transformer for neurodevelopmental disorder diagnosis via direct modeling of real and imaginary components.

Medical image analysis·2025
Same author

Explainable Normative Modeling for Brain Disorder Identification in Resting-State fMRI.

IEEE transactions on medical imaging·2025
Same author

FIESTA: Fourier-Based Semantic Augmentation With Uncertainty Guidance for Enhanced Domain Generalizability in Medical Image Segmentation.

IEEE transactions on neural networks and learning systems·2025
Same author

Virtual reality therapy targeting ideas of reference in patients with psychosis: a single-blind parallel-group randomized controlled trial.

Psychological medicine·2025
Same author

IdenBAT: Disentangled representation learning for identity-preserved brain age transformation.

Artificial intelligence in medicine·2025
Same author

Predictors of relapse after discontinuing antipsychotics in patients with schizophrenia spectrum disorders.

Schizophrenia (Heidelberg, Germany)·2025

Related Experiment Video

Updated: Oct 25, 2025

Author Spotlight: Using Motor Imagery Brain-Computer Interface to Improve Motor and Cognitive Function in Stroke Patients
09:42

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

Published on: September 1, 2023

1.6K

Mutual Information-Driven Subject-Invariant and Class-Relevant Deep Representation Learning in BCI.

Eunjin Jeon, Wonjun Ko, Jee Seok Yoon

    IEEE Transactions on Neural Networks and Learning Systems
    |August 6, 2021
    PubMed
    Summary

    This study introduces a novel framework for brain-computer interfaces (BCIs) that learns subject-invariant features from electroencephalography (EEG) data without adversarial methods. The approach enhances BCI usability by overcoming subject variability and reducing calibration needs.

    More Related Videos

    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

    9.3K
    Author Spotlight: Enhancing Neurorehabilitation Through EEG, Motor Imagery, and Virtual Reality
    10:14

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

    Published on: May 10, 2024

    1.3K

    Related Experiment Videos

    Last Updated: Oct 25, 2025

    Author Spotlight: Using Motor Imagery Brain-Computer Interface to Improve Motor and Cognitive Function in Stroke Patients
    09:42

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

    Published on: September 1, 2023

    1.6K
    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

    9.3K
    Author Spotlight: Enhancing Neurorehabilitation Through EEG, Motor Imagery, and Virtual Reality
    10:14

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

    Published on: May 10, 2024

    1.3K

    Area of Science:

    • Neuroscience
    • Machine Learning
    • Biomedical Engineering

    Background:

    • Deep learning significantly impacts electroencephalography (EEG)-based brain-computer interfaces (BCIs).
    • Subject-specific EEG decoding methods are limited by high variability, requiring extensive calibration and hindering practical application.
    • Transfer learning, particularly domain adaptation, shows promise for BCIs, with adversarial methods demonstrating potential but also risks of negative transfer.

    Purpose of the Study:

    • To propose a novel framework for learning class-relevant and subject-invariant feature representations from EEG data.
    • To address the limitations of adversarial learning in domain adaptation for BCIs, specifically negative transfer.
    • To improve the generalizability and practical usability of EEG-based BCIs.

    Main Methods:

    • A novel deep learning framework utilizing information-theoretic principles, avoiding adversarial learning.
    • Two network components designed to estimate mutual information: one for decomposing features and another for enriching class-discriminative representations.
    • Validation on two large EEG datasets, comparing performance against existing methods and conducting ablation studies.

    Main Results:

    • The proposed framework effectively learns class-relevant and subject-invariant features.
    • Demonstrated superior performance compared to several comparative methods on large EEG datasets.
    • Ablation studies confirmed the contribution of individual network components to the model's effectiveness.

    Conclusions:

    • The novel information-theoretic framework offers a robust solution for learning generalized EEG features, mitigating subject variability.
    • This approach enhances the practical utility of BCIs by reducing reliance on subject-specific calibration.
    • The method provides insights into model decisions and feature distributions, paving the way for more reliable EEG-based BCI systems.