Analysis of absence seizure generation using EEG spatial-temporal regularity measures.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Signal Processing

Background:

• Epileptic seizures are linked to abnormal synchronization in neuronal populations.
• Electroencephalography (EEG) signal dynamics can reveal transitions leading to seizures.
• Understanding seizure onset requires analyzing evolving synchronization patterns.

Purpose of the Study:

• To propose and evaluate a spatial-temporal analysis of EEG using permutation entropy (PE).
• To investigate the transition dynamics preceding absence seizures.
• To compare EEG dynamics in patients with absence seizures and healthy subjects.

Main Methods:

• Spatial-temporal analysis of EEG recordings utilizing permutation entropy (PE).
• Extensive sensitivity analysis of PE performance with respect to parameter settings on scalp EEG.
• Comparison of PE dynamics between 24 patients with absence seizures and 40 healthy subjects.

Main Results:

• Permutation entropy effectively detects different brain states associated with absence seizures.
• Findings suggest a gradual transition model, characteristic of self-organizing networks, may complement or replace the 'jump' transition model for absence seizures.
• Elevated PE levels were consistently observed in frontal-temporal scalp areas, while parieto-occipital areas showed lower PE values in patients, unlike in healthy subjects.

Conclusions:

• The transition to epileptic status appears to be heralded from interictal stages, not just the preictal state.
• Permutation entropy is a valuable tool for quantifying evolving synchronization and detecting brain state changes in EEG.
• The study supports the role of self-organizing network dynamics in the development of absence seizures.