Menu
In the second article in our three-part series on alcohol impairment, we unveil the discovery of a new biomarker: the Optalert Intoxication Scale. Optalert’s algorithm detects alcohol-related driver impairment in real-time, unlocking a new chapter in road safety.
The study administered by KEA Technologies began by measuring participants’ baseline level of impairment before they consumed alcohol. Their BAC and impairment were then monitored across the day until their BAC dropped back to 0.00% (or 8 hours had passed). It emerged that BAC alone does not paint a complete picture of a driver’s ability to operate a vehicle safely.
These findings prompt the question: How should driver monitoring systems detect impairment due to alcohol?
Throughout the study, Optalert collected eyelid data to analyse how blinking behaviour correlates with psychomotor impairment.
The participants were recorded using two cameras placed in the standard in-cabin positions:
Our data scientists processed the videos to extract facial landmarks and generate eyelid closure signals.
They then applied machine learning to develop an advanced algorithm that accurately detects alcohol-related impairment through subtle eyelid movements.
It achieved astonishing predictive accuracy, confirming that eyelid movements reveal when a person is impaired from prior consumption of alcohol, even when they have a low BAC.
Optalert’s data team developed a model solely from the eyelid signals captured by the cameras.
The resulting model achieved an extremely accurate area under curve.
AUC: 0.84
“This discovery is a pivotal step in enhancing road safety,” says Optalert General Manager R&D, Dr. Trefor Morgan.
“Many drivers under the legal limit for BAC are still objectively impaired. We can now detect this in real-time and enact protective countermeasures, many of which are invisible to the driver. The adaptive cruise control could leave more space from the vehicle in front, the lane-keeping assistance can intervene more, and the driver can be encouraged to stop at the next rest stop,” he explains.
“With our eyelid-based impairment algorithm, we can detect the deterioration in cognition that arises after an elevated BAC. This technology will save thousands of lives by mitigating risk before it leads to an accident.”
The Optalert Intoxication Scale – OIS for short – marks a significant advancement in impairment detection, and builds on our Johns Drowsiness Scale (JDS) that detects impairment due to drowsiness.
Unlike BAC-based tests, which measure the amount of alcohol in the bloodstream, Optalert’s eyelid model identifies when a driver is objectively at risk due to impairment, even when at lower BAC levels.
This capability will transform road safety by alerting drivers to their impaired state before risky driving behaviour occurs. In addition, it enables a lot of advanced driver assistance system (ADAS) countermeasures that would be invisible to the driver. These include:
At Optalert, we are committed to advancing driver safety through innovative technology that identifies impairment before it becomes critical. Our innovative eyelid-based algorithms provide a new level of insight into driver alertness, paving the way for safer roads and fewer alcohol-related accidents.