Skip to content

Researchers at Rutgers University in New Jersey say data from smartphone sensors combined with machine learning could detect if someone is under the influence of marijuana.

Researchers set out to develop a proof of concept means to passively detect cannabis use as an alternative to existing detection measures, such as blood, urine or saliva tests. Their results were published in September in the journal Drug and Alcohol Dependence.

“Adverse effects of acute cannabis poisoning have been reported by young adults, with associated consequences such as poor academic and work performance, as well as injuries and deaths from driving while intoxicated with cannabis. “the authors wrote in the study.

The authors conducted an experimental study involving 57 young adults who reported using cannabis at least twice a week. Participants were asked to complete three surveys per day over a 30-day period that asked them how high they felt at any given time, as well as when they had last used cannabis and how much. In total, participants reported 451 episodes of cannabis use.

Participants were also asked to download an app that analyzed GPS data, phone logs, data from the accelerometer and other smartphone sensors, and usage statistics.

By only looking at the time of day, the algorithm was able to accurately detect an episode of cannabis use with 60% accuracy. The data from the smartphone’s sensor alone also produced an accuracy rate of 67%.

However, smartphone sensor data combined with time of day data resulted in an accuracy rate of 90%.

“Using sensors on a person’s phone, we might be able to detect when a person might be intoxicated with cannabis and provide a brief intervention when and where it might have the most. impact to reduce cannabis-related damage, ”said the corresponding author and Rutgers. Professor Tammy Chung in a press release.

GPS data was the most important data set when it came to detecting cannabis use. The researchers found that participants tended to walk shorter distances while they were elevated. The accelerometer data was the second most important feature, as it can be used to measure body movement.

Researchers say this is the first study to examine how smartphone sensors could be used to detect cannabis poisoning.

Chung and his colleagues also participated in a similar 2018 study that investigated whether smartphone data could detect episodes of heavy drinking. In this study, they found that an algorithm measuring smartphone usage patterns, such as screen time, typing speed, and time of day, could detect episodes of consumption. excessive alcohol with an accuracy of 91%.


ctvnews Canada news