Electroencephalographic Resting-State Networks: Source Localization of Microstates

Area of Science:

• Neuroscience
• Brain Imaging
• Computational Neuroscience

Background:

• Elucidating spontaneous brain activity in resting-state networks (RSNs) using electroencephalography (EEG) is challenging due to low signal amplitude and source localization difficulties.
• Electric field topographical analysis offers a method to estimate meta-stable brain states, known as microstates.

Purpose of the Study:

• To reliably estimate electroencephalography resting-state networks (RSNs) by localizing the sources of scalp topographies.
• To identify state-specific brain networks and compare them with existing functional magnetic resonance imaging (fMRI) findings.

Main Methods:

• Utilized k-means clustering to estimate seven resting-state topographies from EEG data (N=164, 256 electrodes).
• Employed a novel source localization method matching sensor and source space temporal patterns for broadband EEG scalp topographies.
• Identified seven state-specific networks by subtracting the mean map from the estimated EEG RSNs.

Main Results:

• Successfully estimated seven EEG RSNs with high reproducibility, demonstrating reliable source localization.
• The mean map highlighted densely connected regions including superior frontal, superior parietal, insula, and anterior cingulate cortices.
• Seven state-specific RSNs were identified, showing partial resemblance and extension to previously identified fMRI-based networks.

Conclusions:

• The developed method reliably estimates EEG RSNs, providing a valuable tool for studying brain dynamics.
• The identified state-specific networks offer new insights into the functional organization of the brain during resting states.
• This approach bridges EEG microstate analysis with RSNs, potentially integrating electrical and hemodynamic measures of brain function.