[Research on Mental Fatigue Detecting Method Based on Sleep Deprivation Models].

Area of Science:

• Neuroscience
• Biomedical Engineering
• Human Factors Engineering

Background:

• Mental fatigue significantly impacts human health, safety, and job performance.
• Dynamic detection of mental fatigue is essential for timely intervention and prevention.
• Electroencephalogram (EEG) signals offer a promising avenue for objective fatigue assessment.

Purpose of the Study:

• To develop and optimize a predictive model for mental fatigue detection using EEG signals.
• To investigate the efficacy of specific EEG features in predicting mental fatigue.
• To determine the minimum number of EEG leads required for accurate fatigue prediction.

Main Methods:

• Induced mental fatigue in subjects through 30 hours of sleep deprivation.
• Extracted key EEG features: relative power, power ratio, center of gravity frequency (CGF), and basic relative power ratio.
• Employed regression analysis to build and optimize a mental fatigue prediction model, including lead selection.

Main Results:

• The initial prediction model achieved a coefficient of determination (R2) of 0.932.
• After lead optimization, a model utilizing only 4 EEG leads demonstrated an R' of 0.811.
• The optimized model's accuracy meets the requirements for daily practical applications of mental fatigue prediction.

Conclusions:

• EEG signal analysis is a viable method for accurate mental fatigue detection.
• A reduced set of 4 EEG leads can provide sufficient accuracy for practical fatigue monitoring.
• The developed model holds potential for improving workplace safety and job performance through fatigue management.