Utilization of a combined EEG/NIRS system to predict driver drowsiness

Area of Science:

• Neuroscience
• Biomedical Engineering
• Transportation Safety

Background:

• Driver drowsiness is a significant cause of automobile accidents globally.
• Existing drowsiness detection methods lack sufficient accuracy, leading to unsatisfactory results.
• There is a critical need for more reliable methods to detect driver fatigue.

Purpose of the Study:

• To introduce a novel approach for driver drowsiness detection using a combination of electroencephalography (EEG) and near-infrared spectroscopy (NIRS).
• To enhance the accuracy of drowsiness detection for improved road safety.
• To identify informative physiological parameters indicative of driver fatigue.

Main Methods:

• Collected electroencephalography (EEG), electrooculography (EOG), electrocardiography (ECG), and near-infrared spectroscopy (NIRS) signals during a simulated driving task.
• Analyzed parameters including blinking rate, eye closure, heart rate, and alpha/beta band power.
• Employed Fisher's linear discriminant analysis for state classification and time series analysis for drowsiness prediction.

Main Results:

• The oxy-hemoglobin concentration change (measured by NIRS) and beta band power (measured by EEG) in the frontal lobe significantly differed between awake and drowsy states.
• These parameters effectively indicated transitions from awake to drowsy states.
• A sharp increase in oxy-hemoglobin concentration and a decrease in beta band power preceded eye closure events.

Conclusions:

• The combination of EEG and NIRS provides a highly accurate method for detecting driver drowsiness.
• Specific changes in frontal lobe oxy-hemoglobin concentration and beta band power are reliable early indicators of fatigue.
• This approach holds significant potential for developing advanced driver assistance systems to prevent fatigue-related accidents.