Design and Implementation of a Drunk Driving Detection System
- Conditions
- Impaired DrivingTransportation AccidentsDrunk DrivingAlcohol Drinking
- Interventions
- Other: Driving under the influence of alcohol with a driving simulator
- Registration Number
- NCT04980846
- Lead Sponsor
- University of Bern
- Brief Summary
To analyse driving behavior of individuals under the influence of alcohol using a validated research driving simulator. Based on the driving variables provided by the simulator the investigators aim at establishing algorithms capable of discriminating sober and drunk driving patterns using machine learning neural networks (deep machine learning classifiers).
- Detailed Description
Driving under the influence of alcohol (or "drunk driving") is one of the most significant causes of traffic accidents. Alcohol consumption impairs neurocognitive and psychomotor function and has been shown to be associated with an increased risk of driving accidents. Automotive technology is highly dynamic, and fully autonomous driving might, in the end, resolve the issue of alcohol impaired accidents. However, autonomous driving (level 4 or 5) is likely to be broadly available only to a substantially later time point than previously thought due to increasing concerns of safety associated with this technology. Therefore, solutions bridging the upcoming period by more rapidly and directly addressing the problem of drunk driving-associated traffic incidents are urgently needed.
On the supposition that driving behaviour differs significantly between sober and drunk states, the investigators assume that different driving patterns in both states can be used to generate drunk driving detection models using machine learning neural networks (deep machine learning classifiers).
Recruitment & Eligibility
- Status
- COMPLETED
- Sex
- All
- Target Recruitment
- 30
- Informed consent as documented by signature.
- In possession of a Swiss or EU driving license for at least two years.
- At least driving 1'000 kilometers per year.
- No special equipment needed when driving.
- Drinks alcohol at least occasionally (moderate/social consumption).
- Fluent in (Swiss) German and no speech impairment.
- Lives in or near Bern.
- Health concerns that are incompatible with alcohol consumption.
- Any potential participant currently taking illegal drugs or medications that interact with alcohol.
- Women who are pregnant or breast feeding.
- Intention to become pregnant during the course of the study.
- Teetotallers (alcohol abstinent persons).
- Alcohol misuse (excessive alcohol consumption habits/risky drinking behaviour (according to WHO definition) and/or PEth in capillary blood > 210 ng/mL at first visit.
- Known or suspected non-compliance or drug abuse.
- Inability to follow the procedures of the study, e.g., due to language problems, psychological disorders, dementia, etc. of the participant.
- Participation in another study with investigational drug within the 30 days preceding and during the present study.
Study & Design
- Study Type
- INTERVENTIONAL
- Study Design
- SINGLE_GROUP
- Arm && Interventions
Group Intervention Description Intervention group Driving under the influence of alcohol with a driving simulator Intervention: Other: Driving under the influence of alcohol with a driving simulator
- Primary Outcome Measures
Name Time Method Accuracy of the DRIVE-model: Diagnostic accuracy of the drunk driving warning system (DRIVE) to detect drunk driving (>= 0.25 mg/l breath alcohol concentration (BrAC)) quantified as the area under the receiver operator characteristics curve (AUC ROC). 480 minutes Accuracy of the DRIVE-model will be assessed using driving data recorded in sober and drunk driving states and driving data will be analysed using applied machine learning technology for impaired driving detection.
- Secondary Outcome Measures
Name Time Method Change of velocity 480 minutes Change of velocity between drunk driving above the legal limit (\>= 0.25 mg/l BrAC), driving within the legal limit (\<0.25 mg/l BrAC) and sober driving.
Change of heart-rate variability 480 minutes Change of heart-rate variability between drunk driving above the legal limit (\>= 0.25 mg/l BrAC), driving within the legal limit (\<0.25 mg/l BrAC) and sober driving.
Change of spinning 480 minutes Change of spinning between drunk driving above the legal limit (\>= 0.25 mg/l BrAC), driving within the legal limit (\<0.25 mg/l BrAC) and sober driving.
Defining the alcohol concentration when driving performance is decreased 480 minutes Breath alcohol concentration (mg/l BrAC) when driving performance begins to be impaired will be assessed based on significantly altered driving parameters for drunk driving above the legal limit (\>= 0.25 mg/l BrAC), below the legal limit (\<0.25 mg/l BAC) and sober driving.
Change of wrist accelerometer 480 minutes Change of wrist accelerometer recorded by a wearable between drunk driving above the legal limit (\>= 0.25 mg/l BrAC), driving within the legal limit (\<0.25 mg/l BrAC) and sober driving.
Change of ethylglucuronide (EtG) in capillary blood 2 weeks Change of ethylglucuronide in capillary blood between before the study day, before driving, during driving in sober state, above the legal limit (\>= 0.25 mg/l BrAC), and under the legal limit (\< 0.25 mg/l BrAC) will be assessed.
Change of steer speed 480 minutes Change of steer speed between drunk driving above the legal limit (\>= 0.25 mg/l BrAC), driving within the legal limit (\<0.25 mg/l BrAC) and sober driving.
Change of time driving over midline 480 minutes Change of time over midline between drunk driving above the legal limit (\>= 0.25 mg/l BrAC), driving within the legal limit (\<0.25 mg/l BrAC) and sober driving.
Change of steer 480 minutes Change of steer between drunk driving above the legal limit (\>= 0.25 mg/l BrAC), driving within the legal limit (\<0.25 mg/l BrAC) and sober driving.
Change of steer torque 480 minutes Change of steer torque between drunk driving above the legal limit (\>= 0.25 mg/l BrAC), driving within the legal limit (\<0.25 mg/l BrAC) and sober driving.
Change of gaze regions of interest 480 minutes Change of gaze regions of interest (as amount and duration of speedometer or front shield inspections) between drunk driving above the legal limit (\>= 0.25 mg/l BrAC), driving within the legal limit (\<0.25 mg/l BrAC) and sober driving.
Change of heart-rate 480 minutes Change of heart-rate between drunk driving above the legal limit (\>= 0.25 mg/l BrAC), driving within the legal limit (\<0.25 mg/l BrAC) and sober driving.
Diagnostic accuracy in detecting drunk driving (>= 0.25 mg/l BrAC) quantified as the area under the receiver operator curve (AUC-ROC) using audio data 480 minutes Using audio data recorded by microphones accuracy in drunk driving detection will be analysed with applied machine learning technology.
Change of brake 480 minutes Change of brake between drunk driving above the legal limit (\>= 0.25 mg/l BrAC), driving within the legal limit (\<0.25 mg/l BrAC) and sober driving.
Change of swerving 480 minutes Change of swerving between drunk driving above the legal limit (\>= 0.25 mg/l BrAC), driving within the legal limit (\<0.25 mg/l BrAC) and sober driving.
Change of gaze behavior 480 minutes Change of gaze behavior (as gaze velocity, acceleration or direction changes) between drunk driving above the legal limit (\>= 0.25 mg/l BrAC), driving within the legal limit (\<0.25 mg/l BrAC) and sober driving.
Driving performance while being sober, above and within the legal limit 480 minutes Based on significantly altered driving parameters above the legal limit of drunk driving (\>= 0.25 mg/l breath alcohol concentration in Switzerland) driving performance while being sober, above and within the legal limit will be assessed.
Change of electrodermal activity (EDA) 480 minutes Change of EDA between drunk driving above the legal limit (\>= 0.25 mg/l BrAC), driving within the legal limit (\<0.25 mg/l BrAC) and sober driving.
Change of skin temperature 480 minutes Change of skin temperature between drunk driving above the legal limit (\>= 0.25 mg/l BrAC), driving within the legal limit (\<0.25 mg/l BrAC) and sober driving.
Change of cortisol in capillary blood 2 weeks Change of cortisol in capillary blood between before the study day, before driving, during driving in sober state, above the legal limit (\>= 0.25 mg/l BrAC), and under the legal limit (\< 0.25 mg/l BrAC) will be assessed.
Accuracy-comparison of DRIVE-model and DRIVEplus-model 480 minutes Diagnostic accuracy of the drunk driving warning system (DRIVING) to detect drunk driving (\>= 0.25 mg/l BrAC) quantified as the area under the receiver operator characteristics curve (AUC ROC) using only driving parameters (DRIVE-model) will be compared to the DRIVE-model with additional integration of physiological parameters (heart-rate, heart-rate variability, electrodermal activity (EDA), skin temperature, accelerometer, facial expression, gaze behavior, and radar) (DRIVEplus-model)
Change of gaze events 480 minutes Change of gaze events (as amount and duration of fixations or saccades) between drunk driving above the legal limit (\>= 0.25 mg/l BrAC), driving within the legal limit (\<0.25 mg/l BrAC) and sober driving.
Change of oral fluid cortisol 2 weeks Change of oral fluid cortisol between before the study day, before driving, during driving in sober state, above the legal limit (\>= 0.25 mg/l BrAC), and under the legal limit (\< 0.25 mg/l BrAC) will be assessed.
Change of ethylsulfate (EtS) in capillary blood 2 weeks Change of ethylsulfate in capillary blood between before the study day, before driving, during driving in sober state, above the legal limit (\>= 0.25 mg/l BrAC), and under the legal limit (\< 0.25 mg/l BrAC) will be assessed.
Change of phosphatidylethanol (PEth) in capillary blood 2 weeks Change of phosphatidylethanol in capillary blood between before the study day, before driving, during driving in sober state, above the legal limit (\>= 0.25 mg/l BrAC), and under the legal limit (\< 0.25 mg/l BrAC) will be assessed.
Diagnostic accuracy in detecting drunk driving (>= 0.25 mg/l BrAC) quantified as the area under the receiver operator characteristics curve using physiological data 480 minutes Accuracy of drunk driving detection using physiological data (heart-rate, heart-rate variability, skin temperature, EDA, accelerometer) recorded with wearable devices during the study period will be analysed using applied machine learning technology.
Diagnostic accuracy in detecting drunk driving (>= 0.25 mg/l BrAC) quantified as the area under the receiver operator curve (AUC-ROC) using video data 480 minutes Using video data recorded by a camera and a thermal camera accuracy in drunk driving detection will be analysed with applied machine learning technology.
Self-estimation of alcohol concentrations 480 minutes Correlation between self-estimated alcohol concentrations and measured breath alcohol concentrations will be assessed.
Diagnostic accuracy in detecting drunk driving (>= 0.25 mg/l BrAC) quantified as the area under the receiver operator curve (AUC-ROC) using eye-tracking data 480 minutes Using eye-tracking data recorded by a camera and an eye-tracker (to record gaze behaviour) accuracy in drunk driving detection will be analysed with applied machine learning technology.
Self-estimation of driving performance 480 minutes Correlation between self-estimated driving performance and measured driving performance based on significantly altered driving parameters in sober state, above the legal limit (\>= 0.25 mg/l BrAC), and under the legal limit (\< 0.25 mg/l BrAC) will be assessed. Driving performance will be assessed with two questionnaires. Driving performance will be assessed with on an absolute 5-point scale from 0-5 (a higher value means higher driving performance).
Self-estimation of workload 480 minutes Correlation between self-estimated workload and measured driving performance based on significantly altered driving parameters in sober state, above the legal limit (\>= 0.25 mg/l BrAC), and under the legal limit (\< 0.25 mg/l BrAC) will be assessed. Workload will be assessed an absolute 20-point scale from 0-20 (a higher value means higher workload).
Self-estimation of sleepiness levels 480 minutes Correlation between self-estimated sleepiness levels, measured driving performance, and measured breath alcohol concentrations based on significantly altered driving parameters in sober state, above the legal limit (\>= 0.25 mg/l BrAC), and under the legal limit (\< 0.25 mg/l BrAC) will be assessed. Sleepiness will be assessed on an absolute 7-point scale from 0-6 (a lower value means less sleepiness).
Diagnostic accuracy in detecting drunk driving (>= 0.25 mg/l BrAC) quantified as the area under the receiver operator curve (AUC-ROC) using radar sensor data 480 minutes Using radar sensor data (directed on the body of the driver) accuracy in drunk driving detection will be analysed with applied machine learning technology.
Incidence of Adverse Events (AEs) 3 weeks Adverse Events will be recorded at each study visit.
Incidence of Serious Adverse Events (SAEs) 3 weeks Serious Adverse Events will be recorded at each study visit.
Trial Locations
- Locations (1)
University of Bern
🇨ðŸ‡Bern, Switzerland