Epileptic seizure onset correlates with long term EEG functional brain network properties

Area of Science:

• Neuroscience
• Epilepsy Research
• Computational Neuroscience

Background:

• Epilepsy is a neurological disorder characterized by recurrent seizures.
• Understanding the temporal dynamics of brain networks is crucial for predicting seizure onset.
• Long-term electroencephalography (EEG) provides valuable data for analyzing brain network properties.

Purpose of the Study:

• To investigate the correlation between long-term EEG functional brain network properties and epileptic seizure onset times.
• To identify periodic temporal structures within brain network properties and their relationship to seizure occurrence.

Main Methods:

• Constructed binary functional brain networks from long-term, multichannel EEG data of nine epilepsy patients.
• Quantified network properties, including average network degree, network efficiency, and clustering coefficient.
• Utilized circular statistics to correlate the phases of network periodicities with seizure onset times.

Main Results:

• Network properties like degree, efficiency, and clustering coefficient showed significant temporal fluctuations.
• Identified consistent periodic temporal structures (1.5-24 hour periods) in brain network properties across subjects.
• Found that the instantaneous phases of 3.5hr, 5.5hr, 12hr, and 24hr network degree periodic components were not uniformly distributed, indicating a relationship with seizure generation.

Conclusions:

• Functional brain network properties exhibit periodic temporal structures that are related to seizure generation.
• The phase of these network periodicities may serve as a potential biomarker for seizure occurrence.
• This study provides insights into the dynamic interplay between brain network function and epilepsy.