Directed differential connectivity graph of interictal epileptiform discharges

Area of Science:

• Neuroscience
• Computational Biology
• Medical Engineering

Background:

• Accurate localization of epileptic regions is crucial for effective epilepsy treatment.
• Interictal events in focal epilepsy can provide insights into seizure onset zones.
• Current methods for seizure localization may have limitations in precision.

Purpose of the Study:

• To develop and validate a quantitative method for localizing leading epileptic regions using interictal data.
• To assess the utility of graph analysis of interictal electroencephalography (iEEG) for predicting seizure onset.
• To improve the prediction of seizure onset zones in focal epilepsy.

Main Methods:

• Utilized wavelet transform and cross-correlation coefficient to analyze temporal couplings in iEEG data.
• Developed a differential connectivity graph (DCG) to highlight significant changes between epileptic and non-epileptic states.
• Employed mutual information and multiobjective optimization for localizing epileptic regions within the DCG.

Main Results:

• The proposed differential connectivity graph (DCG) method successfully identified significant connections related to interictal events.
• Postprocessing techniques effectively localized potential leading epileptic regions.
• The method demonstrated good performance when compared to visual inspection and electrically stimulated seizures in five epilepsy patients.

Conclusions:

• Quantitative graph analysis of interictal iEEG data can effectively localize leading epileptic regions.
• The developed differential connectivity graph (DCG) approach shows promise for predicting seizure onset zones.
• This method offers a valuable tool for improving epilepsy surgery planning and patient outcomes.