OCT 01, 2021 6:30 AM PDT

Smartphone Technology Could Help Predict Cannabis Intoxication

WRITTEN BY: Ryan Vingum

Cannabis use can affect the body in a range of ways, including slowing response time. A slower response time affects our ability to function in a range of contexts, from lower-stakes contexts, like work, to higher-stakes/more dangerous contexts, like driving. Like with alcohol, finding ways to detect when someone is under the influence is important. However, existing methods for detecting whether someone is under the influence of cannabis, such as certain blood and saliva tests, have their limitations, including taking longer to determine results

Researchers at multiple institutions are studying how to use smartphone technology and algorithmic thinking as a minimally invasive way to detect when someone might be under the influence of cannabis, enabling the delivery interventions to prevent people from operating under the influence in high-risk situations.

The study, published in Drug and Alcohol Dependence, included participants ages 18-25 who reported regular cannabis usage (at least twice a week). Researchers worked to collect a range of data that could be configured with a type of machine learning to predict levels of intoxication. Data included phone surveys 3 times a day, participant reports of when they started using cannabis each day, and their perceived level of intoxication (on a scale of 1-10, with 10 being the most intoxicated). 

Researchers used a machine learning algorithm, called Light Gradient Boosting Machine (LGBM), to analyze the collected data and make predictions about, or “detect,” when participants may have self-reported “high” levels of intoxication, which was accurate about 60% of the time. Using smartphone sensors (such as those that power a GPS or accelerometer) increased the accuracy of these predictions to 90%.

Some researchers are attempting to develop other methods for detecting intoxication similar to alcohol, such as breathalyzers. However, cannabis affects the body in very different ways compared to alcohol, suggesting that different solutions are needed to accurately and effectively test if someone is under the influence and how being under the influence impairs their ability to function. According to this research, though, we may be heading in a promising direction.  

Sources: Science Daily; McGill; Drug and Alcohol Dependence; Neural Information Processing Systems

About the Author
  • Science writer and editor, with a focus on simplifying complex information about health, medicine, technology, and clinical drug development for a general audience.
You May Also Like
JUL 05, 2021
Cannabis Sciences
Prenatal Cannabis Exposure Lessens Effects of Antidepressants
JUL 05, 2021
Prenatal Cannabis Exposure Lessens Effects of Antidepressants
Researchers at Indiana University have found that mice exposed to cannabinoids while in the womb are less responsive to ...
JUL 24, 2021
Cannabis Sciences
Causal Link Between Cannabis Use Disorder and Schizophrenia
JUL 24, 2021
Causal Link Between Cannabis Use Disorder and Schizophrenia
A study from Denmark has found that the proportion of schizophrenia cases linked to cannabis use disorder has increased ...
AUG 28, 2021
Cannabis Sciences
Could Eye-tracking Data Detect THC Levels?
AUG 28, 2021
Could Eye-tracking Data Detect THC Levels?
Eye-tracking data shows promise for detecting levels of tetrahydrocannabinol (THC). The corresponding study was publishe ...
SEP 30, 2021
Cannabis Sciences
Concerning New Evidence Regarding Cannabis Use During Pregnancy
SEP 30, 2021
Concerning New Evidence Regarding Cannabis Use During Pregnancy
According to data from a National Vital Statistics Report published in March 2021, the percentage of infants born with a ...
OCT 08, 2021
Cannabis Sciences
The Synergistic Mystery of the Entourage Effect
OCT 08, 2021
The Synergistic Mystery of the Entourage Effect
The entourage effect is a much talked about characteristic attributed with the unique psychoactive effects associated wi ...
OCT 15, 2021
Health & Medicine
Anti-seizure properties found in the Cannabinoid CBGA
OCT 15, 2021
Anti-seizure properties found in the Cannabinoid CBGA
A growing body of research suggests cannabis’s effectiveness in minimizing seizure activity, but a recent study hi ...
Loading Comments...