Extracting information from cortical connectivity patterns estimated from high resolution EEG recordings: a...

Area of Science:

• Neuroscience
• Computational Neuroscience
• Graph Theory Applications

Background:

• High-resolution electroencephalography (EEG) enables precise estimation of cortical activity and causal relationships between brain regions.
• Interpreting functional brain network connectivity patterns objectively across subjects and groups remains a challenge.
• Graph theory offers tools to analyze complex network architectures, previously applied to social networks, the web, and proteomics.

Purpose of the Study:

• To demonstrate the suitability of applying graph theory to analyze functional brain network architecture derived from EEG.
• To investigate differences in network architecture between healthy subjects and spinal cord injured patients during motor tasks.
• To compare network architecture between two distinct motor tasks (foot-lip vs. foot movement) in healthy subjects.

Main Methods:

• Functional brain networks were constructed from non-invasive EEG recordings.
• Graph theory indexes were used to analyze the 'architecture' of these functional connectivity networks.
• Two experiments were conducted: comparing healthy vs. spinal cord injured groups, and comparing two motor tasks in healthy individuals.

Main Results:

• Functional connectivity networks derived from EEG exhibited ordered properties.
• Significant differences were found between the network architectures of the two subject groups and the two motor tasks.
• The observed network properties differed significantly from those of random networks with similar characteristics.

Conclusions:

• Graph theory provides a robust framework for objectively analyzing and comparing functional brain network architectures.
• This approach can differentiate between neurological conditions (e.g., spinal cord injury) and task-specific brain activity patterns.
• The graph theory method is applicable to cerebral connectivity patterns derived from EEG and potentially other neuroimaging techniques.