A method to quantify invariant information in depth-recorded epileptic seizures

Area of Science:

• Neuroscience
• Computational Biology
• Epilepsy Research

Background:

• Stereoelectroencephalography (SEEG) is crucial for understanding brain structure interactions during seizures.
• Analyzing SEEG signals in temporal lobe epilepsy (TLE) patients offers insights into seizure dynamics.

Purpose of the Study:

• To develop and apply a novel methodology for comparing SEEG seizure recordings.
• To objectively identify reproducible seizure mechanisms and spatio-temporal signatures in TLE.

Main Methods:

• A three-step process: signal segmentation, segment characterization/labeling, and vector-based comparison of observations.
• Utilizing a vectorial extension of the Wagner-Fischer algorithm for sequence alignment and signature extraction.
• Matching non-equal duration SEEG observations via vector deformation under minimal cost.

Main Results:

• Objective, reproducible results on seizure mechanisms in four TLE patients.
• Extraction of invariant spatio-temporal signatures (ictal patterns) from SEEG signals without prior assumptions.
• Identified signatures, including spikes and wave patterns, reflect reproducible propagation schemes.

Conclusions:

• The methodology provides objective insights into seizure dynamics and reproducible propagation patterns.
• Extracted spatio-temporal signatures aid in understanding the underlying epileptogenic networks.
• This approach enhances the analysis of SEEG data for epilepsy research.