How machine learning could curb the opioid crisis

As a nationwide opioid crisis continues to consume the U.S., one pharmacist suggests we take advantage of recent advances in health tech—namely AI—to fight rising death rates and falling life expectancies.

Writing in Specialty Pharmacy Times, Lee Feigert, PharmD, a consultant pharmacist for a PACE program in Pennsylvania, said there’s “no clear end in sight” to the opioid epidemic, which costs more than 115 Americans their lives each day. U.S. life expectancy dropped for a second consecutive year in 2017—the first time that had happened since the ’60s—and prescription opioid abuse alone costs the country upwards of $78.5 billion a year.

A handful of medical tech companies are leveraging AI and machine learning to better predict opioid relapses and overdoses, Feigert said. Chicago firm Triggr Health, which has spent the past five years training a machine learning algorithm to predict recovery relapses, connects addicts to their care teams through a phone app that analyzes their texting patterns, location and sleep history.

“The goal of this monitoring is to identify patterns that indicate a stronger likelihood of relapse,” Feigert wrote. “Information gathered related to these patterns is combined with data such as client drug history and keywords such as ‘craving’ from communications between Triggr Health users and staff, in hopes of identifying whether an individual is at risk of relapse.”

Another company, Pittsburgh-based startup Behaivior, is working to collect data through a wearable device that scrutinizes heart rate, skin temperature and heart rate variability to assess whether a patient is in a pre-relapse state of craving.

“Less talk and more tangible solutions are needed, such as those offered by Triggr and hopefully by Behaivior in the near future,” Feigert wrote.

Read the full article below: