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

Global search metaheuristics for neural mass model calibration.

Imaging neuroscience (Cambridge, Mass.)·2026

Same author

The microstructure-weighted human connectome: network properties and structure-function correlations across spatial scales.

bioRxiv : the preprint server for biology·2026

Same author

Dynamics-informed priors (DIP) for neural mass modelling.

Imaging neuroscience (Cambridge, Mass.)·2026

Same author

Clinical Manifestations.

Alzheimer's & dementia : the journal of the Alzheimer's Association·2025

Same author

Evaluating the dependence of ADC-fMRI on haemodynamics in breath-hold and resting-state conditions.

Imaging neuroscience (Cambridge, Mass.)·2025

Same author

Apparent Diffusion Coefficient fMRI shines light on white matter resting-state connectivity compared to BOLD.

Communications biology·2025

Same journal

Lifespan Trajectories of the Brain's Functional Complexity Characterized by Multiscale Sample Entropy.

NeuroImage·2026

Same journal

Pleasant fragrance modulates dyadic social sharing of positive emotion: Sharer-centered socioemotional enhancement effect and its neural couplings.

NeuroImage·2026

Same journal

Altered Functional Hierarchical and Sequential Organization in Individuals with Schizophrenia during Auditory Processing.

NeuroImage·2026

Same journal

Mechanical Deformation Explains Distinct Neuroimaging Patterns and Etiologies in Brain Trauma.

NeuroImage·2026

Same journal

Ventral striatum temporal interference brain stimulation enhances the reward-positivity event-related potential and reduces anxiety.

NeuroImage·2026

Same journal

NeuroHarm‑Kit: An Open‑Source Toolbox for Benchmarking Deep‑Learning Harmonization of Multi‑Site T1‑Weighted MRI.

NeuroImage·2026

See all related articles

Search research articles

Related Experiment Video

Updated: Sep 23, 2025

Dynamic Inter-subject Functional Connectivity Reveals Moment-to-Moment Brain Network Configurations Driven by Continuous or Communication Paradigms

Dynamic Inter-subject Functional Connectivity Reveals Moment-to-Moment Brain Network Configurations Driven by Continuous or Communication Paradigms

Published on: March 21, 2019

Using deep clustering to improve fMRI dynamic functional connectivity analysis.

Arthur P C Spencer¹, Marc Goodfellow²

¹Clinical Research and Imaging Centre, University of Bristol, Bristol, UK; Translational Health Sciences, Bristol Medical School, University of Bristol, Bristol, UK.

|May 13, 2022

Summary

This summary is machine-generated.

Deep autoencoders improve dynamic functional connectivity (dFC) state analysis by enhancing k-means clustering performance on resting-state fMRI data, especially with heterogeneous subject data.

Keywords:

Autoencoders Deep learning Dimensionality reduction Dynamic functional connectivity Sliding window correlations

More Related Videos

Using Informational Connectivity to Measure the Synchronous Emergence of fMRI Multi-voxel Information Across Time

Using Informational Connectivity to Measure the Synchronous Emergence of fMRI Multi-voxel Information Across Time

Published on: July 1, 2014

Co-analysis of Brain Structure and Function using fMRI and Diffusion-weighted Imaging

Co-analysis of Brain Structure and Function using fMRI and Diffusion-weighted Imaging

Published on: November 8, 2012

Related Experiment Videos

Last Updated: Sep 23, 2025

Dynamic Inter-subject Functional Connectivity Reveals Moment-to-Moment Brain Network Configurations Driven by Continuous or Communication Paradigms

Dynamic Inter-subject Functional Connectivity Reveals Moment-to-Moment Brain Network Configurations Driven by Continuous or Communication Paradigms

Published on: March 21, 2019

Using Informational Connectivity to Measure the Synchronous Emergence of fMRI Multi-voxel Information Across Time

Using Informational Connectivity to Measure the Synchronous Emergence of fMRI Multi-voxel Information Across Time

Published on: July 1, 2014

Co-analysis of Brain Structure and Function using fMRI and Diffusion-weighted Imaging

Co-analysis of Brain Structure and Function using fMRI and Diffusion-weighted Imaging

Published on: November 8, 2012

Area of Science:

Neuroimaging
Computational Neuroscience
Data Science

Background:

Dynamic functional connectivity (dFC) analysis of resting-state fMRI commonly uses sliding-window correlations (SWC) and k-means clustering.
K-means performance is sensitive to window parameters and signal-to-noise ratio, impacting dFC state temporal property accuracy.
Subject heterogeneity can further compromise group-level clustering, potentially leading to erroneous conclusions in clinical vs. control comparisons.

Purpose of the Study:

To quantify k-means ability to estimate dFC state temporal properties in multi-subject cohorts.
To explore methods for maximizing clustering performance in dFC analysis.
To investigate the utility of dimensionality reduction techniques preceding k-means clustering for improved dFC state identification.

Main Methods:

Explored deep autoencoders for dimensionality reduction before k-means clustering (deep clustering).
Compared deep clustering against PCA, UMAP, and direct k-means (L1/L2) on synthetic and real-world fMRI data.
Evaluated clustering performance using synthetic datasets simulating heterogeneous subject data.

Main Results:

Deep clustering demonstrated superior performance in capturing temporal properties of dFC states on synthetic data compared to other methods.
PCA, UMAP, and direct k-means approaches were often insufficient for accurate dFC state temporal property estimation.
Application to real-world human fMRI data confirmed that the choice of dimensionality reduction significantly impacts group-level dFC state measurements.

Conclusions:

Deep autoencoder-based dimensionality reduction enhances k-means clustering for dFC state analysis.
This deep clustering approach offers improved accuracy for estimating dFC state temporal properties, particularly in heterogeneous cohorts.
The selection of dimensionality reduction significantly influences findings in group-level dFC analyses.