Quantitative Analysis of Inter- and Intrahemispheric Coherence on Epileptic Electroencephalography Signal

Area of Science:

• Neuroscience
• Biomedical Engineering
• Epilepsy Research

Background:

• Epileptic seizures cause dynamic changes in neuronal activity and brain region connectivity.
• Electroencephalography (EEG) coherence quantifies brain region connectivity by measuring statistical correlations between scalp electrodes.
• Previous coherence analyses examined all EEG electrodes, overlooking seizure-specific active regions.

Purpose of the Study:

• To quantitatively analyze inter- and intrahemispheric coherence in epileptic EEG signals.
• To investigate the correlation between channel activity and coherence changes during seizures.
• To identify brain area connectivity changes during epileptic seizures using an energy-based channel selection.

Main Methods:

• Utilized EEG data from ten patients in the CHB-MIT dataset.
• Applied an energy-based channel selection to focus on the most active brain regions during seizures.
• Calculated pair-wise electrode spectral coherence in the full band and five sub-bands, correlating with channel energy.

Main Results:

• Observed a significant decrease in EEG coherence (Coh) in the most active channels during seizures (ictal) compared to preictal periods.
• This decrease was particularly pronounced in lower EEG sub-bands, showing a strong correlation with increased channel activity.
• Reduced coherence (Coh < Coh) indicates diminished neuronal connectivity in high-activity regions during epileptic events.

Conclusions:

• Epileptic EEG signal connectivity is effectively analyzed by focusing on the most active brain regions.
• A strong correlation exists between decreased spectral coherence and heightened channel activity during seizures.
• The findings suggest that reduced neuronal connectivity is a key characteristic of epileptic conditions in active brain areas.