Related Experiment Video
Updated: May 7, 2026
Co-analysis of Brain Structure and Function using fMRI and Diffusion-weighted Imaging
Published on: November 8, 2012
Mapping the voxel-wise effective connectome in resting state FMRI
Guo-Rong Wu1, Sebastiano Stramaglia, Huafu Chen
1Faculty of Psychology and Educational Sciences, Department of Data Analysis, Ghent University, Ghent, Belgium ; Key Laboratory for NeuroInformation of Ministry of Education, School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu, China.
This study introduces a new method to map brain networks using Granger causality density (GCD) at the voxel level. This approach reveals brain hubs and information flow, offering new insights into brain function and organization.
Area of Science:
- Neuroscience
- Network Science
- Computational Biology
Background:
- Functional magnetic resonance imaging (fMRI) is widely used to study brain function via functional connectivity.
- Investigating directed connections and network dynamics at the voxel level remains challenging.
- Existing methods often struggle with high-dimensional and short fMRI datasets.
Purpose of the Study:
- To develop a novel multivariate Granger causality approach for reconstructing voxel-wise effective connectivity networks.
- To analyze the topological organization of these dynamical directed brain networks.
- To identify voxel-wise hubs of information flow and influence.
Main Methods:
- Integration of information theory and dynamical network architecture for variable selection.
- Aggregation of conditional information sets based on community organization.
- Application of Granger causality analysis to high-dimensional, short fMRI time series data.
Main Results:
- First depiction of voxel-wise Granger causality density (GCD) representing incoming and outgoing information.
- Identification of brain hubs based on GCD, showing analogies and differences with functional and anatomical connectomes.
- Demonstration of feasibility in studying directed network architecture and identifying hubs at the voxel level.
Conclusions:
- The novel approach enables a new description of global brain organization and information influence.
- It allows for the study of directed network architecture at the voxel level.
- This method can identify critical hubs within the brain's dynamic network structure.
