The cannabis industry is booming, but the use of this plant has been linked to negative side effects such as addiction and impaired driving. Can smartphone sensors detect when someone is high?
According to a recent research from the Rutgers Institute for Health, Health Care Policy, and Aging Research, a smartphone sensor, similar to what is used in GPS systems, may be used to detect whether or not someone is inebriated after ingesting marijuana.
According to a research published in Drug and Alcohol Dependence, a combination of temporal characteristics (monitoring the time of day and day of week) and smartphone sensor data showed a 90% accuracy rate in identifying episodes of cannabis intoxication in the natural environment.
“We might be able to detect when a person is experiencing cannabis intoxication and deliver a brief intervention when and where it would have the most impact to reduce cannabis-related harm using the sensors in a person’s phone,” said corresponding author Tammy Chung, professor of psychiatry and director of the Center for Population Behavioral Health at the Rutgers Institute for Health.
Intoxication with cannabis has been linked to delayed reaction times, which may impact performance at work or school, as well as impaired driving conduct, which can lead to injuries or deaths. Existing screening methods, such as blood, urine, and saliva tests, have limitations as markers of cannabis intoxication and impairment in everyday life.
The researchers looked at daily data from young people who said they used cannabis at least twice a week. They looked at phone surveys, self-reported cannabis use, and continuous phone sensor data to see how important time of day and week are in identifying usage, as well as which phone sensors are best at detecting self-reported cannabis intoxication.
They discovered that time of day and day of week had 60% accuracy in identifying self-reporting of cannabis intoxication, while a combination of time characteristics and smartphone sensor data had 90% accuracy.
The most significant phone sensor characteristics for detecting self-reported cannabis intoxication were travel patterns using GPS data – at times when individuals reported feeling high – and movement data from an accelerometer that identifies various movements.
Researchers utilized low-burden techniques to identify intoxication in everyday life (monitoring time of day and week, and analyzing phone sensor data), and discovered that utilizing phone sensors to detect subjective intoxication from cannabis use is very feasible.
Future study should look at how well the system performs in identifying intoxicated vs non-intoxicated reports in people who don’t use cannabis very often. Researchers should look at reports of intoxication using methods that law enforcement could employ to see whether there is a stronger link between self-reported cannabis usage and drunkenness.
Stevens Institute of Technology, Stanford University, Carnegie Mellon University, University of Tokyo, Japan, and University of Washington, Seattle are among the study’s authors.