Study Shows Smartphone Sensor Data Could Detect Cannabis Exposure
As companies continue to experiment with 'cannabis testing' and attempt to perfect the technology to correctly detect the presence of cannabis, a new report shows how data from phone sensors (like GPS) can be used as detector. A new report shows how data from sensors in phones, like GPS devices, can be used to detect.
A smartphone sensor, similar to the one used in GPS systems, could be a way to determine whether or not a person is under the influence of cannabis after using marijuana, according to one. new study of the Rutger Institute which bases its research on health, care policies and aging.
According to the study, published in Drug and Alcohol Dependence, which assessed the feasibility of using data from smartphone sensors to identify episodes of cannabis poisoning in the natural environment, a combination of temporal characteristics (followed by time of day and day of the week) and data from smartphone sensors achieved an accuracy rate of 90%.
Using sensors on a person's phone, we might be able to detect when a person is "getting drunk" with cannabis and offer them an alternative when and where they are most likely to be. safe to reduce the potential harm from cannabis, ”said author Tammy Chung, professor of psychiatry and director of the Center for Population Behavioral Health.
Drunkenness or "intoxication" with cannabis has been determined by slower reaction time, loss of certain performance at work or school and also impaired driving behavior which could lead to accidents. Existing measures such as blood, urine or saliva tests have limitations as a reliable indicator of cannabis consumption and do not give any value with regard to the possible lowering of the faculties linked to cannabis in everyday life. .
The researchers analyzed daily data collected from young adults who reported using cannabis at least twice a week. They examined telephone surveys, spontaneous reports of cannabis use, and data from continuous telephone sensors to determine the importance of the time of day and day of the week in detecting use and identifying telephone sensors. most useful for detecting declared cannabis poisoning.
They found that the day and time of consumption was 60% accurate in detecting overconsumption of cannabis, according to stakeholder statements, and that the combination of time characteristics and data from smart phone sensors had 90% accuracy in detection.
From the GPS data, the movement patterns at the times when they reported feeling under the effects of cannabis, and the movement data from the accelerometer that detects the various movements, were the most prominent characteristics of the phone's sensors.
The researchers used light-restricting methods (tracking the time of day and day of the week and analyzing data from the phone's sensor) to detect intoxication in everyday life and found that the feasibility of the use of the phone's sensors to detect subjective intoxication due to cannabis use is strong.
Future research should study the performance of the algorithm to rank the data of degrees of influence of consumption versus those not in people who use cannabis less frequently. Research should investigate these reports using tools law enforcement could use, showing a stronger correlation with self-reported cannabis use.
The authors of the study are professors at Stevens Institute of Technology, Stanford University, Carnegie Mellon University, the University of Tokyo (Japan) and the University of Washington (Seattle).