Driver Fatigue Detection Using Heart Rate Variability Features from 2-Minute Electrocardiogram Signals While...

Area of Science:

• Physiology
• Transportation Safety
• Biomedical Engineering

Background:

• Driver fatigue is a significant cause of road accidents and fatalities.
• Heart rate variability (HRV) analysis offers a potential method for monitoring driver alertness.
• Existing methods often require longer physiological signal recordings.

Purpose of the Study:

• To investigate the efficacy of heart rate variability (HRV) features for detecting driver fatigue.
• To explore the impact of sex differences on HRV-based fatigue detection.
• To develop a rapid driver fatigue detection system using short electrocardiogram (ECG) signals.

Main Methods:

• Simulated driving experiments were conducted with college students.
• Electrocardiogram (ECG) signals were recorded for 2-minute intervals to derive HRV features.
• Statistical analysis (Mann-Whitney U test) identified significant HRV differences between alert and fatigued states.
• Decision tree models were employed for fatigue classification, analyzing data for all subjects, males, and females separately.

Main Results:

• Significant differences in HRV features were observed between alert and fatigued states across all subjects, males, and females (p < 0.01).
• Fatigue detection models achieved high classification accuracies: 86.3% for all subjects, 94.8% for males, and 92.0% for females.
• Sex-specific models demonstrated superior performance compared to a general model.

Conclusions:

• Sex differences significantly influence HRV features related to driver mental states.
• Incorporating sex-specific analysis enhances the precision of HRV-based driver fatigue detection systems.
• Utilizing 2-minute ECG signals enables faster and more efficient driver fatigue monitoring compared to traditional 5-minute methods.