Multifractal spatiotemporal dynamics in human epileptiform stereoelectroencephalography recordings

Area of Science:

• Neuroscience
• Complex Systems Analysis
• Epilepsy Research

Background:

• Non-stationary time series data, common in biological systems, requires advanced analytical frameworks.
• Multifractal formalism provides a method to analyze nonlinear, scale-invariant features across multiple time scales.
• Understanding the spatiotemporal dynamics of epileptic seizures is crucial for improving patient outcomes.

Purpose of the Study:

• To investigate multifractal features of spatiotemporal correlations in seizure activity using stereoelectroencephalography (sEEG) data.
• To apply multifractal detrended fluctuation analysis (MFDFA) to sEEG recordings from epilepsy patients.
• To explore the potential of MF-derived metrics for understanding seizure onset, evolution, and network involvement.

Main Methods:

• Multifractal detrended fluctuation analysis (MFDFA) was applied to sEEG recordings from five patients with refractory focal temporal epilepsy.
• Analysis encompassed pre-ictal, ictal, and post-ictal states, as well as different anatomical brain regions.
• MFDFA-derived metrics were statistically analyzed to identify unique features and temporal variations.

Main Results:

• Evidence for a multifractal architecture in sEEG-recorded epileptiform signals in vivo was reported for the first time.
• MFDFA features captured altered spatiotemporal trends across seizure states and brain regions.
• Increased fluctuations in MF-derived metrics were observed in resected temporal lobe structures compared to non-resected networks.

Conclusions:

• MFDFA metrics offer a potentially valuable tool for visualizing, quantifying, and interpreting network involvement in seizure activity.
• The findings highlight the importance of investigating high-complexity dynamics in intracranial sEEG recordings.
• These insights may contribute to improved surgical decision-making for patients with medically intractable epilepsy.