SEEG-based epileptic seizure network modeling and analysis for pre-surgery evaluation.

Area of Science:

• Neurosurgery
• Computational Neuroscience
• Epileptology

Background:

• Refractory epilepsy often necessitates surgical intervention to manage seizures.
• Stereo-electroencephalography (SEEG) is a key diagnostic tool for localizing seizure origins.
• Accurate identification of epileptogenic zones is crucial for successful epilepsy surgery.

Purpose of the Study:

• To develop and validate a computational methodology for analyzing SEEG data.
• To assist clinicians in recommending optimal surgical targets for focal epilepsy.
• To characterize seizure networks and propagation pathways using graph theory.

Main Methods:

• A seizure network (graph) model was developed to represent spatial and temporal ictal dynamics.
• Nodes represented epileptogenic regions, and edges depicted propagation pathways weighted by directed transfer function.
• K-means clustering was applied to group network nodes and identify target surgical areas.

Main Results:

• The proposed methodology was applied to analyze SEEG data from ten focal epilepsy patients.
• The seizure network model effectively characterized ictal event distribution and propagation.
• Consistent recommendations were observed between the computational method, clinical decisions, and surgical outcomes.

Conclusions:

• The developed seizure network analysis provides a valuable tool for surgical planning in focal epilepsy.
• This computational approach can enhance the precision of identifying target areas for epilepsy surgery.
• The findings suggest improved surgical decision-making and potentially better patient outcomes through data-driven analysis.