Related Experiment Video
Updated: Dec 31, 2025
08:51
Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms
Published on: November 1, 2019
6.0K
Comparing connectivity metrics in cortico-cortical evoked potentials using synthetic cortical response patterns
David Prime1, Matthew Woolfe1, David Rowlands2
1Griffith University School of Engineering and Built Environment, Nathan, QLD, Australia; Mater Advanced Epilepsy Unit, Brisbane, QLD, Australia.
Journal of Neuroscience Methods
|January 14, 2020
Summary
This study compared metrics for analyzing Cortico-Cortical Evoked Potentials (CCEPs) in epilepsy patients. Root Mean Square (RMS) proved most sensitive for detecting brain connectivity patterns, validating current methods.
Area of Science:
- Neuroscience
- Epileptology
- Signal Processing
Background:
- Cortico-Cortical Evoked Potentials (CCEPs) are a novel brain mapping technique in epilepsy investigations.
- Current CCEP analysis lacks a comparative study of different connectivity metrics.
Purpose of the Study:
- To compare the efficacy of various metrics in analyzing CCEP data.
- To identify the most sensitive metric for detecting simulated neural patterns.
Main Methods:
- Developed a novel method by superimposing synthetic cortical responses onto SEEG data.
- Compared two standard CCEP metrics against eight time series similarity metrics (TSSMs).
Main Results:
- Root Mean Square (RMS) demonstrated high sensitivity across diverse patterns, though less so for epileptiform ones.
- Autoregressive (AR) coefficients were highly sensitive to epileptiform patterns.
- Elastic warping metrics showed lower sensitivity to simulated responses.
Conclusions:
- RMS is a robust and sensitive metric for CCEP analysis, supporting existing SEEG-CCEP literature.
- AR coefficients show potential for investigating epileptic networks.
