Transfer Function between EEG and BOLD Signals of Epileptic Activity

Area of Science:

• Neuroscience
• Medical Imaging
• Epilepsy Research

Background:

• Simultaneous electroencephalogram (EEG)-functional Magnetic Resonance Imaging (fMRI) is increasingly used for epilepsy evaluation.
• Current EEG-fMRI analysis models epileptic events as boxcar signals, limiting hemodynamic complexity.
• Modeling the EEG-BOLD transfer function remains an open challenge in epilepsy research.

Purpose of the Study:

• To propose a novel methodology for integrating simultaneous EEG-fMRI data in epilepsy analysis.
• To incorporate a transfer function from EEG to Blood Oxygen Level Dependent (BOLD) signals for improved modeling.
• To identify brain areas associated with epileptic activity using this new approach.

Main Methods:

• Independent Component Analysis (ICA) of EEG data.
• Extraction of EEG-BOLD transfer function metrics from ICA time courses.
• Prediction of fMRI data and identification of brain areas linked to EEG epileptic activity.

Main Results:

• The proposed methodology identified plausible, consistent, and more significant brain activations compared to conventional methods.
• Frequency-weighted EEG metrics outperformed power-weighted metrics, aligning with prior research.
• The method was tested on ictal and interictal EEG-fMRI recordings from a patient with hypothalamic hamartoma.

Conclusions:

• The developed methodology offers a more flexible and potentially more accurate way to analyze simultaneous EEG-fMRI data in epilepsy.
• Frequency-weighted metrics are crucial for effectively modeling EEG-BOLD dynamics.
• Further validation with larger patient groups is needed to confirm reproducibility, specificity, and sensitivity.