Visualization and modelling of STLmax topographic brain activity maps

Area of Science:

• Neuroscience
• Dynamical Systems Theory
• Biomedical Engineering

Background:

• Epileptic seizures are characterized by abnormal brain activity.
• Identifying the seizure onset zone is crucial for understanding and treating epilepsy.
• Previous work suggested a link between Short-Term Maximum Lyapunov Exponent (STLmax) topography and seizure onset zones.

Purpose of the Study:

• To model the spatial distribution of STLmax from EEG data.
• To automatically extract quantitative features of STLmax topography.
• To investigate the relationship between STLmax spatial distribution and epileptic seizure onset mechanisms.

Main Methods:

• Processed one-hour preictal EEG segments from four patients (two scalp, two intracranial).
• Estimated Short-Term Maximum Lyapunov Exponent (STLmax) profiles.
• Modeled spatial STLmax maps using a combination of two Gaussian functions.

Main Results:

• The fitted Gaussian model allowed automatic extraction of quantitative information on STLmax spatial distribution.
• Low STLmax levels were observed in the seizure origin zone prior to seizure onset in 3 out of 4 patients.
• EEG power topographic maps did not yield useful information for seizure onset localization.

Conclusions:

• STLmax topography modeling shows potential as a tool for monitoring epileptic brain dynamics.
• The findings suggest that STLmax spatial distribution can provide insights into seizure onset mechanisms.
• Further research with more elaborate approaches is needed to improve method specificity.