Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

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

Artificial Intelligence-Mediated Discharge Document for Accessible Health Care (AIM-HEALTH): Protocol for a Prospective, Observational, Noninterventional Study.

JMIR research protocols·2026

Same author

Symbiotic brain-machine drawing via visual brain-computer interfaces.

npj biomedical innovations·2026

Same author

NeuroFM: Toward Precision Neuroimaging with Foundation Models for Individualized Brain Health Estimation.

medRxiv : the preprint server for health sciences·2026

Same author

Generalizable neural models of emotional engagement and disengagement.

Social cognitive and affective neuroscience·2026

Same author

Correction: When randomisation is not good enough: Matching groups in intervention studies.

Psychonomic bulletin & review·2026

Same author

Delineating In Vivo T1-Weighted Intensity Profiles Within the Human Insula Cortex Using 7-Tesla MRI.

Human brain mapping·2026

Same journal

Development and Standardization of a Simple Zebrafish Larval Model of Global Hypoxia-Reoxygenation: Recapitulating Key Pathological Features Associated with Cerebral Ischemia-Reperfusion Injury.

Journal of neuroscience methods·2026

Same journal

Metal Neurotoxicity assessment: a new imaging and analysis pipeline for primary neurons in co-cultures.

Journal of neuroscience methods·2026

Same journal

Corrigendum to "Human peripheral nerve xenografts and rat peripheral nerve allografts implanted to the striatum: Methodology and initial findings of cell-based therapy" [J. Neurosci. Methods 434 (2026) 110825].

Journal of neuroscience methods·2026

Same journal

Sombor-based graph-theoretic framework for the structural characterization of neuro-metabolic organic acids.

Journal of neuroscience methods·2026

Same journal

Pupil-DLC: An open-source deep learning pipeline for scalable, marker-less tracking of pupil dynamics across conscious and unconscious states.

Journal of neuroscience methods·2026

Same journal

Time as the language of Behavior: events, sequences, patterns and meanings.

Journal of neuroscience methods·2026

See all related articles

Search research articles

Related Experiment Video

Updated: Jan 6, 2026

Training Dogs for Awake, Unrestrained Functional Magnetic Resonance Imaging

Training Dogs for Awake, Unrestrained Functional Magnetic Resonance Imaging

Published on: October 13, 2019

Transfer learning of deep neural network representations for fMRI decoding.

Michele Svanera¹, Mattia Savardi², Sergio Benini²

¹Department of Information Engineering, University of Brescia, Italy; Institute of Neuroscience and Psychology, University of Glasgow, UK.

Journal of Neuroscience Methods

|October 5, 2019

Summary

This summary is machine-generated.

This study introduces a novel method to decode brain activity from fMRI data by transferring deep learning features. This approach significantly improves decoding performance compared to using imaging data alone.

Keywords:

Brain decoding Convolutional Neural Network Deep learning MultiVoxel Pattern Analysis Transfer learning fMRI

More Related Videos

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

Decoding Natural Behavior from Neuroethological Embedding

Decoding Natural Behavior from Neuroethological Embedding

Published on: October 3, 2025

Related Experiment Videos

Last Updated: Jan 6, 2026

Training Dogs for Awake, Unrestrained Functional Magnetic Resonance Imaging

Training Dogs for Awake, Unrestrained Functional Magnetic Resonance Imaging

Published on: October 13, 2019

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

Decoding Natural Behavior from Neuroethological Embedding

Decoding Natural Behavior from Neuroethological Embedding

Published on: October 3, 2025

Area of Science:

Neuroscience
Machine Learning
Computer Vision

Background:

Deep neural networks, particularly Convolutional Neural Networks (CNNs), excel at object classification but face challenges in brain imaging due to limited data.
Functional Magnetic Resonance Imaging (fMRI) data scarcity hinders direct application of CNNs for decoding subject states or perceptions.

Purpose of the Study:

To develop a robust method for transferring information from deep learning (DL) features to fMRI data for improved decoding.
To establish a multivariate link between fMRI data and the fully connected layer (fc7) of a CNN for enhanced analysis.

Main Methods:

Utilized Reduced Rank Regression with Ridge Regularisation to connect fMRI data with CNN's fc7 layer.
Applied object image classification using reconstructed fc7 features on two fMRI datasets (movie clips and static images).

Main Results:

Demonstrated significant reconstruction of fc7 features from fMRI data.
Achieved significant decoding performance using the reconstructed fc7 features.

Conclusions:

The proposed method enhances fMRI-based decoding by mapping functional data to CNN features.
Benefits include unsupervised extraction of stimuli representations and dimensionality reduction for high-dimensional neuroimaging data.