A novel real-time driving fatigue detection system based on wireless dry EEG

Area of Science:

• Neuroscience
• Biomedical Engineering
• Transportation Safety

Background:

• Sustaining attention is critical in high-stakes fields like transportation and security.
• Existing methods for mental fatigue detection require improvement for real-time applications.
• Electroencephalographic (EEG) signals offer a promising avenue for objective fatigue monitoring.

Purpose of the Study:

• To develop a novel, real-time driving fatigue detection methodology.
• To utilize dry Electroencephalographic (EEG) signals for monitoring mental fatigue.
• To create an integrated metric for predicting the degree of driving fatigue.

Main Methods:

• Employed power spectrum density (PSD) and sample entropy (SE) for online fatigue detection.
• Utilized wavelet packets transform (WPT) to analyze specific EEG frequency bands (theta, alpha, beta).
• Developed fatigue-related indexes and fused them into an integrated metric, calibrated per subject.

Main Results:

• The proposed methods demonstrated effectiveness in real-time fatigue detection.
• Fatigue prediction using EEG signals showed consistency with behavioral measures (reaction time).
• Individualized calibration improved the performance of the fatigue detection system.

Conclusions:

• The developed EEG-based methodology provides an effective approach for real-time driving fatigue detection.
• This technique has significant potential applications in transportation safety and other attention-critical domains.
• The integration of PSD and SE analysis offers a robust measure of mental fatigue.