Functional Complex Networks Based on Operational Architectonics: Application on EEG-based Brain-computer Interface...

Area of Science:

• Neuroscience
• Computational Neuroscience
• Network Science

Background:

• Understanding complex brain dynamics is crucial for advancing neuroscience.
• Existing methods for functional brain graph generation have limitations.
• Integrating Operational Architectonics (OA) with Network Neuroscience offers a new perspective.

Purpose of the Study:

• To present a novel computational method for analyzing brain complex dynamics and states.
• To construct functional brain graphs using EEG signals and OA principles.
• To evaluate the method's performance in classification tasks for brain-computer interfaces.

Main Methods:

• Utilizes Operational Architectonics (OA) for non-parametric EEG signal segmentation to identify Rapid Transition Processes (RTPs).
• Generates undirected weighted complex networks based on RTP time coordinates, ensuring scale-free topology.
• Estimates brain connectivity network metrics to form feature vectors for machine learning.

Main Results:

• Successfully applied the method to an EEG-based Brain-Computer Interface (BCI) dataset.
• Achieved classification accuracies above chance level using a Naïve Bayes classifier for imagery pronunciation tasks.
• Demonstrated competitive performance compared to other state-of-the-art functional network generation approaches.

Conclusions:

• The developed method provides a robust framework for analyzing brain complex dynamics and states.
• It offers a valuable tool for neuroscientists to enhance brain research algorithms.
• The approach shows promise for applications in brain-computer interfaces and predictive modeling.