Multiplex Limited Penetrable Horizontal Visibility Graph from EEG Signals for Driver Fatigue Detection

Area of Science:

• Neuroscience
• Network Science
• Transportation Safety

Background:

• Driver fatigue is a major cause of road accidents.
• Current methods for fatigue detection require further investigation into underlying mechanisms.
• Understanding brain fatigue behavior is crucial for developing effective countermeasures.

Purpose of the Study:

• To develop a novel method for detecting driver fatigue and understanding brain fatigue behavior.
• To apply the Multiplex Limited Penetrable Horizontal Visibility Graph (Multiplex LPHVG) method to electroencephalogram (EEG) data.
• To differentiate between alert and mental fatigue states using network analysis.

Main Methods:

• Constructed brain networks from EEG signals using the Multiplex LPHVG method.
• Analyzed network topology using clustering coefficient, global efficiency, and characteristic path length.
• Combined network measures with average edge overlap for state classification.

Main Results:

• The Multiplex LPHVG method achieved high accuracy in classifying alert and fatigue driving states from EEG signals.
• A significant increase in clustering coefficient was observed as the brain transitioned from an alert to a fatigued state.
• The study successfully demonstrated the method's efficacy for fatigue detection and provided insights into brain fatigue.

Conclusions:

• The novel Multiplex LPHVG method is effective for driver fatigue detection using EEG signals.
• The findings offer new insights into brain behavior associated with fatigue driving.
• This approach advances the understanding and detection of mental fatigue.