Related Experiment Video
Updated: May 1, 2026
07:12
Using Informational Connectivity to Measure the Synchronous Emergence of fMRI Multi-voxel Information Across Time
Published on: July 1, 2014
12.3K
Estimation of direct nonlinear effective connectivity using information theory and multilayer perceptron
Ali Khadem1, Gholam-Ali Hossein-Zadeh2
1Control and Intelligent Processing Center of Excellence, School of Electrical and Computer Engineering, University College of Engineering, University of Tehran, Tehran, Iran.
Journal of Neuroscience Methods
|April 23, 2014
Summary
A new method, βmRMR-MLP-GC, accurately estimates direct nonlinear causal brain networks in high-dimensional data. This approach surpasses existing techniques in sensitivity and specificity for analyzing neural signals like EEG.
Area of Science:
- Neuroscience
- Computational Neuroscience
- Data Science
Background:
- Quantifying direct nonlinear causal couplings in high-dimensional datasets remains a challenge.
- Existing effective connectivity measures often fail to capture complex nonlinear interactions or are unsuitable for large-scale neural data.
Purpose of the Study:
- To introduce a novel method, βmRMR-MLP-GC, for estimating direct nonlinear effective connectivity in high-dimensional neural datasets.
- To address the limitations of current methods in analyzing complex causal relationships within neural signals.
Main Methods:
- Utilized βmRMR for optimal feature selection of neural signal regressors.
- Employed a multilayer perceptron (MLP) with cross-validation for multivariate signal characterization.
- Defined a Granger Causality (GC)-based measure to quantify causal interactions among neural channels.
Main Results:
- Achieved >95% sensitivity and specificity on simulated high-dimensional datasets with diverse structures.
- Identified significant posterior-to-anterior brain activity propagation in resting-state EEG alpha band.
- Demonstrated superior performance compared to Granger Causality Index, Conditional Granger Causality Index, and Transfer Entropy.
Conclusions:
- βmRMR-MLP-GC offers a robust tool for estimating direct nonlinear causal networks in high-dimensional data.
- The method accurately identifies reproducible information flow and highlights dominant brain regions in neural activity.
- This advancement facilitates a deeper understanding of brain connectivity.
