Related Experiment Video
Updated: Dec 29, 2025
14:27
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
Published on: June 26, 2013
16.1K
Exploring the Correlation Between M/EEG Source-Space and fMRI Networks at Rest.
Jennifer Rizkallah1,2, Hassan Amoud3, Matteo Fraschini4
1Univ Rennes, LTSI, 35000, Rennes, France. jennifer.rizkallah.jr@gmail.com.
Brain Topography
|January 31, 2020
Summary
Choosing the right functional connectivity (FC) method is crucial when comparing magneto/electro-encephalography (M/EEG) and fMRI brain networks. Phase Locking Value (PLV) and Amplitude Envelope Correlation (AEC) showed significant correlations with fMRI networks.
Area of Science:
- Neuroscience
- Brain Imaging
- Network Science
Background:
- Magneto/electro-encephalography (M/EEG) source connectivity offers high temporal and spatial resolution for brain network estimation.
- Comparing M/EEG-derived networks with functional Magnetic Resonance Imaging (fMRI) networks is essential for multimodal brain research.
- Various functional connectivity (FC) methods exist, differing in their handling of zero-lag connections, which may impact network comparability.
Purpose of the Study:
- To evaluate the influence of different FC methods on the correlation between M/EEG source-space and fMRI resting-state brain networks.
- To compare FC methods that include zero-lag connectivity (PLV, AEC) against those that exclude it (PLI, orthogonalized PLV/AEC).
- To determine which FC methods yield M/EEG source-space networks most comparable to established fMRI networks.
Main Methods:
- Resting-state M/EEG data from 74 healthy participants were analyzed.
- Two families of FC methods were tested: those including zero-lag (PLV, AEC) and those removing it (PLI, orthogonalized PLV/AEC).
- M/EEG-derived networks were compared with fMRI networks from the Human Connectome Project.
Main Results:
- All tested FC methods showed low correlations between M/EEG source-space and fMRI networks.
- Phase Locking Value (PLV) and Amplitude Envelope Correlation (AEC) demonstrated statistically significant correlations with fMRI networks (ρ=0.12 and ρ=0.06, respectively).
- Other FC methods, including Phase Lag Index (PLI) and orthogonalized approaches, did not yield significant correlations.
Conclusions:
- The selection of FC methods significantly impacts the comparability of M/EEG source-space and fMRI brain networks.
- PLV and AEC are more suitable for aligning M/EEG source-space networks with fMRI networks at rest compared to other methods.
- Further research involving simulations and real data is needed to refine understanding of M/EEG-fMRI network relationships.
