A NEW AI model could potentially detect whether someone has recently used cannabis.
The advance could one day help medical professionals quickly determine whether someone in need of medical care has used cannabis.
Currently, professionals use urine, saliva or a strand of hair to test cannabis use.
However, it may take a few days for these laboratory results to be analyzed and available.
This led Stevens Institute of Technology researcher Sang Won Bae and a team of colleagues to look for a way to quickly determine whether someone might be using cannabis.
For their research, they examined 33 adults who consumed cannabis at least twice a week.
They had participants report their consumption every day for up to 30 days and also wore an activity tracker.
This tracker collected information such as heart rate, step count and sleep quality.
Other sensors on participants’ phones provided data on their micro-movements.
Specifically, this meant the way they held their phone, which was crucial to their stability and coordination.
Data collected from some of these participants was then fed to an AI to determine whether anyone may have consumed cannabis.
The researchers then used the AI on the data from the remaining participants to test its performance.
Their results showed that the AI had 85 percent accuracy in detecting people who had been moderately high in the last five minutes.
However, some experts remain skeptical of the results and believe further research is needed.
According to New Scientist, Joseph Wu from Stanford University in California and Mark Chandy from Western University in Canada both agree that a larger group of subjects needs to be tested.
Additionally, some experts expressed concerns about the reliability of consumers in reporting their cannabis use.