Over six months, more than 60 people receiving medication-assisted treatment for opioid use disorder answered surveys on their smartphones about their mental health, psychological state and environment — three times a day.

When deep learning models — a form of artificial intelligence — analyzed their responses, the technology forecasted relapse risk and likelihood of continuing treatment exceptionally well, according to a study published in the Journal of Substance Use and Addiction Treatment in June.

The National Institutes of Health-funded study — which included researchers from New Hampshire, New York, California, Massachusetts and Maryland — found AI using real-time monitoring has the potential to serve as a strong predictive tool and early-warning system. It could pave the way f

See Full Page