Detecting functional hubs of ictogenic networks

Area of Science:

• Neuroscience
• Epileptology
• Computational Neuroscience

Background:

• Quantitative EEG (qEEG) challenges the traditional hyper-synchrony model of epileptic seizures, suggesting complex functional network re-organization.
• Despite advancements, qEEG is underutilized in clinical decision-making for epilepsy management.

Purpose of the Study:

• To investigate intracranial EEG signal dynamics by analyzing functional networks and information flow.
• To identify seizure generation hubs and their predictive value for surgical outcomes in pharmaco-resistant epilepsy.

Main Methods:

• Analysis of 198 seizures from 27 patients using transfer entropy to quantify information flow in functional networks.
• Identification of in-, out-, and total-hubs representing maximal information flow.
• Correlation of hub locations with surgical outcomes in patients who underwent resection.

Main Results:

• Seizure hubs predominantly occurred around the middle of seizures, not at the beginning or end.
• A higher percentage of out- or total-hubs in the resected area correlated with good postoperative outcomes (p=0.01 for out-hubs, p=0.04 for total-hubs).
• In-hub locations did not show clear predictive value for surgical success.

Conclusions:

• Functional network analysis using qEEG can identify critical brain areas for seizure generation.
• This approach may guide targeted interventions and improve surgical planning for epilepsy.
• Hub dynamics provide insights into seizure generation beyond traditional EEG interpretation.