Skip to main content
KBS_Icon_questionmark link-ico
doctors-looking-at-mri-images-2023-11-27-05-03-08-utc ;

Unlocking the Dynamic Brain: meet DySCo, a general framework for Dynamic Functional Connectivity

Giuseppe De Alteriis

PhD Student at Neuroimaging, King's IoPPN

14 April 2025

We’re excited to introduce DySCo, a powerful framework to analyse dynamic brain data with a set of guidelines and code implemented in MATLAB and Python. DySCo is designed to enhance how we analyse and understand brain connectivity based on imaging results. Here’s a closer look at what makes this work significant, as published in PLOS Computational Biology.

The challenges in today’s functional connectivity field

Functional connectivity is an essential tool in fMRI research, as it helps uncover the complex interactions between different brain regions. However, current approaches come with several limitations. Traditional methods often provide only static snapshots, capturing isolated moments of brain activity while missing the transient nature of functional connectivity. Additionally, computational efficiency remains a challenge—many existing techniques are not optimized to handle large datasets, such as imaging data, which might have a high number of pixels/voxels. Another major issue is limited flexibility, as many frameworks struggle to adapt across different imaging or experimental conditions.

What does DySCo address?

DySCo is a dynamic functional connectivity framework designed to overcome these limitations by offering a more comprehensive approach to studying brain connectivity and its evolution over time. Unlike static models, DySCo captures the continuous evolution of functional connectivity, allowing researchers to analyze how brain networks fluctuate. It enhances sensitivity by detecting subtle, transient connectivity patterns that may otherwise go unnoticed. Most importantly, DySCo is the first general framework that integrates and unifies multiple previous attempts at dynamic functional connectivity, making it a versatile and adaptable tool for a wide range of research applications.

Where can DySCo be Applied?

This framework has broad applications across neuroscience, including:

  • Resting-State and Task-Related Studies – DySCo enables researchers to track changes in functional connectivity during both passive and active brain states.
  • Neurological and Psychiatric Conditions – Its ability to detect dynamic changes makes it useful for studying disorders such as epilepsy, Alzheimer’s disease, and other neurodegenerative or psychiatric conditions.

Why choose DySCo over traditional frameworks?

DySCo offers several key advantages over conventional approaches. By providing dynamic insights, it captures the flow in time of neural interactions, moving beyond the static limitations of traditional models. The framework is also optimized for performance, ensuring a more accurate and computationally efficient representation of brain activity. Additionally, DySCo is designed for user-friendly integration, with clear documentation and accessible code, making it easy for researchers to incorporate into their existing workflows.

How to adopt DySCo for your MRI research

If you’re interested in applying DySCo to your MRI studies, getting started is simple. The paper includes detailed documentation to guide researchers through the integration process. Ready-to-use scripts and examples make it easier than ever to explore dynamic functional connectivity in your own work. Plus, DySCo fosters a collaborative community, allowing researchers to engage, refine, and expand upon its applications.

Beyond MRI: expanding to other imaging modalities

DySCo is a general framework which is designed, potentially, for any neural signal. The framework can also be applied to EEG (electroencephalography) and MEG (magnetoencephalography). Moreover, its adaptable nature means it could be extended to other neuroimaging techniques, paving the way for new research opportunities across various domains.

Join the DySCo community!

You can find our original article DySCo: A general framework for dynamic functional connectivity here.

If you are interested in applying the method, you can find the MATLAB and Python software here.

If you have any question, feedback, or want to collaborate with us, please email Giuseppe.de_alteriis@kcl.ac.uk or oliver.sherwood@kcl.ac.uk. We are very interested in extending the method to other types of functional neural data!

Latest news