It's no secret that not getting enough sleep can cause a person to be a drowsy, dangerous driver. What's difficult, however, is being able to objectively determine whether or not a driver is sleep-deprived. Soon, a blood test could provide the answer.
Led by Professor Derk-Jan Dijk, a team at the University of Surrey's Sleep Research Centre recently conducted an experiment in which 36 test subjects went one night without sleep. Blood samples were taken from the participants throughout a 40-hour period which included that night, allowing the expression levels of thousands of genes to be measured.
After the data was analyzed using a machine learning algorithm, it was found that consistent changes in the expression levels of 68 specific genes could be used to reliably differentiate between blood samples drawn from sleep-deprived and well-rested individuals – the technology currently has an accuracy rate of 92 percent.
"Identifying these biomarkers is the first step to developing a test which can accurately calculate how much sleep an individual has had," says U Surrey's Prof. Simon Archer. "The very existence of such biomarkers in the blood after only a period of 24-hour wakefulness shows the physiological impact a lack of sleep can have on our body."
Ultimately, it is hoped that the test could be used by companies to check if their drivers are fit for work, or to determine if sleep deprivation played a part in a car accident. It might also be used in the assessment of sleep disorders.
A paper on the research was recently published in the journal SLEEP.
Source: University of Surrey via EurekAlert
Then add in the warrentless search issues and what we have is nothing useful. I would say legislators would never fall for this, except that history teaches a sad lesson on that front. Right now, I'm more in favor of fully autonomous vehicles than ever.
Declaring "Accuracy" tends to be misleading as it doesn't define what that means (parameters are important)...
So what exactly does it mean if the markers are not present??
When one analyses the data even such a high "Accuracy" may result in orders of magnitude greater error rates....
Statistically, statistics and lies may be interchangeable. lol
Another chink in the armour of workers rights ( to work), behold your robo-overlords... "Bob isn't fit to drive today so we'll just file an Autonomous routing plan." - solong as Bob gets a paid rest day it may be seen as ok.