Researchers at Rutgers University in New Jersey say smartphone sensor data combined with machine learning can detect whether a person is under the influence of marijuana.
Researchers set out to develop a proof-of-concept method to passively detect cannabis use as an alternative to existing detection measures such as blood, urine or saliva tests. Their findings were published in September in Journal Drug and Alcohol Dependence.
“Adverse effects of acute cannabis intoxication by young adults have been reported, such as poor academic and work performance, and injuries and fatalities due to driving while ‘high’ on cannabis,” the authors write in the study.
The authors conducted a study experiment involving 57 young adults who reported using cannabis at least twice a week. Participants were asked to complete three surveys a day over a 30-day period that asked how high they felt at a given time, as well as when they last reported the amount of cannabis and consumption. was used. Overall, participants reported 451 episodes of cannabis use.
Participants were also asked to download an app that analyzed GPS data, phone logs, accelerometer data and other smartphone sensor and usage statistics.
When looking at the time of day only, the algorithm was able to accurately detect an episode of cannabis use with 60 percent accuracy. Smartphone sensor data alone was also able to produce an accuracy rate of 67 percent.
However, combining smartphone sensor data with day-to-day data yielded an accuracy rate of 90 percent.
Corresponding author and Rutgers said, “By using sensors in a person’s phone, we may be able to detect when a person is experiencing cannabis intoxication and when and where it may have the greatest impact.” ” Professor Tammy Chung in a news release.
GPS data was the most important dataset when it came to tracing cannabis use. The researchers found that participants would travel shorter distances when they were higher. Accelerometer data was the second most important feature, as it can be used to measure body movements.
Researchers say this is the first study to look at how smartphone sensors can be used to detect cannabis intoxication.
Chung and his colleagues were also involved in Similar study from 2018 It examined whether smartphone data could detect episodes of heavy drinking. In that study, they found that an algorithm that measured smartphone-use patterns, such as screen-on duration, typing speed and time of day, could detect episodes of heavy drinking with 91 percent accuracy.