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

BrainDEC: An M/EEG Unsupervised Representation Learning Framework with Disentangled Equivariance Constraint.

Interdisciplinary sciences, computational life sciences·2026
Same author

Explainable Model Differencing for MEG Decoding via Predict-Probability Differences and Boundary-Optimized Rules.

Annals of the New York Academy of Sciences·2026
Same author

Magnetoencephalography Decoding Transfer Approach: From Deep Learning Models to Intrinsically Interpretable Models.

IEEE journal of biomedical and health informatics·2024
Same author

A Novel 3D Reversible Data Hiding Scheme Based on Integer-Reversible Krawtchouk Transform for IoMT.

Sensors (Basel, Switzerland)·2023
Same author

An Efficient Routing Protocol for Quantum Key Distribution Networks.

Entropy (Basel, Switzerland)·2022
Same author

Clinical Effect of Emergency Dermabrasion Combined with Biological Dressing A on Wound Microcirculation and Preventing Sepsis in Deep Degree-II Burns.

Emergency medicine international·2022

Related Experiment Video

Updated: Aug 3, 2025

Detecting Pre-Stimulus Source-Level Effects on Object Perception with Magnetoencephalography
09:25

Detecting Pre-Stimulus Source-Level Effects on Object Perception with Magnetoencephalography

Published on: July 26, 2019

6.9K

A Model-Agnostic Feature Attribution Approach to Magnetoencephalography Predictions Based on Shapley Value.

Yongdong Fan, Haokun Mao, Qiong Li

    IEEE Journal of Biomedical and Health Informatics
    |April 7, 2023
    PubMed
    Summary
    This summary is machine-generated.

    This study introduces a new method for interpreting deep learning models in magnetoencephalography (MEG) decoding. The approach enhances trust and practical use by explaining individual predictions and identifying key brain signal features.

    More Related Videos

    Author Spotlight: IntelliSleepScorer — A High-Accuracy, Accessible GUI Software for Automated Sleep Stage Scoring in Mice and its Application in Psychiatric Research
    04:54

    Author Spotlight: IntelliSleepScorer — A High-Accuracy, Accessible GUI Software for Automated Sleep Stage Scoring in Mice and its Application in Psychiatric Research

    Published on: November 8, 2024

    583
    How to Detect Amygdala Activity with Magnetoencephalography using Source Imaging
    10:48

    How to Detect Amygdala Activity with Magnetoencephalography using Source Imaging

    Published on: June 3, 2013

    22.3K

    Related Experiment Videos

    Last Updated: Aug 3, 2025

    Detecting Pre-Stimulus Source-Level Effects on Object Perception with Magnetoencephalography
    09:25

    Detecting Pre-Stimulus Source-Level Effects on Object Perception with Magnetoencephalography

    Published on: July 26, 2019

    6.9K
    Author Spotlight: IntelliSleepScorer — A High-Accuracy, Accessible GUI Software for Automated Sleep Stage Scoring in Mice and its Application in Psychiatric Research
    04:54

    Author Spotlight: IntelliSleepScorer — A High-Accuracy, Accessible GUI Software for Automated Sleep Stage Scoring in Mice and its Application in Psychiatric Research

    Published on: November 8, 2024

    583
    How to Detect Amygdala Activity with Magnetoencephalography using Source Imaging
    10:48

    How to Detect Amygdala Activity with Magnetoencephalography using Source Imaging

    Published on: June 3, 2013

    22.3K

    Area of Science:

    • Neuroscience
    • Machine Learning
    • Biomedical Engineering

    Background:

    • Deep learning significantly improves magnetoencephalography (MEG) decoding performance.
    • Lack of interpretability in deep learning hinders practical application and user trust in MEG decoding.

    Purpose of the Study:

    • To develop a feature attribution method for interpretable deep learning-based MEG decoding.
    • To provide interpretative support for individual MEG predictions, addressing a critical gap in current algorithms.

    Main Methods:

    • Transformed MEG samples into feature sets.
    • Utilized modified Shapley values to assign contribution weights to features.
    • Optimized Shapley values via reference sample filtering and antithetic sample pair generation.

    Main Results:

    • Achieved high attribution accuracy with an Area Under the Deletion test Curve (AUDC) of 0.005.
    • Validated feature importance against neurophysiological theories through visualization.
    • Demonstrated signal compression to 1/16th size with minimal (<0.19%) performance loss.

    Conclusions:

    • The proposed feature attribution approach enhances interpretability in deep learning-based MEG decoding.
    • The method is model-agnostic, applicable to diverse decoding models and brain-computer interface (BCI) applications.
    • Increased interpretability can foster compliance and end-user trust in advanced neuroimaging analysis.