Could artificial intelligence be the spark that will end the opioid epidemic? | TechCrunch

The opioid epidemic has been of incredible complexity, perplexing researchers for much of the past two decades as they attempt to better understand the evolution of the social and systemic factors that drive people to start abusing opioids and also to identify potential overdoses. staines.

These woefully tedious and often flawed efforts all occur as doctors work to provide safe and effective treatments and other resources to those in the throes of addiction.

As both researchers and clinicians examine the widespread and persistent reach of the opioid epidemic, they are now curiously exploring artificial intelligence and asking: Could this be the shot on the moon that will end the opioid epidemic?

The healthcare sector is not one to jump on the bandwagon, notoriously slow to pilot and implement new technologies. And this trend is not without consequences. One report suggests that the industry loses more than $8.3 billion annually due to the late adoption or non-adoption of technologies such as advanced electronic health records.

But the toll of opioid epidemics is greater than what is on record. Back in 1999, over 1 million people died from drug-related overdoses. In 2021, 106,699 overdose deaths occurred in America, one of the highest per capita deaths in the country’s history. About 75% of all of these overdoses were attributable to opioid use, including painkillers such as Vicodin and Percocet, as well as street drugs such as heroin.

Though the Centers for Disease Control and Prevention and the National Institutes of Health have poured billions of dollars into awareness, education and prescription tracking programs, the epidemic has remained stubbornly persistent.

For the past decade, I have conducted research on the opioid epidemic in rural and urban communities across America, including New York City and rural southern Illinois.

Most in my field agree, albeit reluctantly, that there is an incredible amount of guesswork involved in identifying the complex risks drug users face. What drugs will they receive? Will they inject them, sniff them, or smoke them? Who, if anyone, will they use if they overdose and need help?

That’s not it. Professionals also regularly battle idiosyncratic federal and state guidelines about effective treatments for opioid use disorder, such as suboxone. And they also find themselves catching up with increasingly unpredictable drug supplies tainted with cheap synthetic opioids like fentanyl, which is largely responsible for the recent surge in opioid-related overdose deaths.

While AI developments like ChatGPT have been what captured the imagination of most of the public, public health researchers and biomedical engineers have been quietly masterminding a revolution in AI-powered medicine, with prevention and addiction treatment the new beneficiaries.

Innovations in this area primarily use machine learning to identify individuals who may be at risk of developing opioid use disorder, disengaging from treatment, and relapse. For example, researchers at the Georgia Institute of Technology recently developed machine learning techniques to effectively identify individuals on Reddit who were at risk for fentanyl abuse, while other researchers developed a tool to spot misinformation about treatments for fentanyl. opioid use disorder, both of which could allow colleagues and advocates to intervene with education.

Other AI-powered programs, such as Sobergrid, are developing the ability to detect when individuals are at risk of relapse, for example, based on their proximity to bars, then linking them to a recovery counselor.

The most impactful developments concern the reduction of overdoses, often caused by drug mixing. At Purdue University, researchers have developed and tested a wearable device that can detect signs of an overdose and automatically inject an individual with naloxone, an overdose-reversing agent. Another crucial development has been the creation of tools to detect dangerous contaminants in drug supplies, which could radically reduce fentanyl overdoses.

Despite this immense promise, is there any question that facial recognition technology could be used to spot people who appear stoned, leading to discrimination and abuse? Uber has already taken this kind of capability a step further in 2008, by trying to patent a technology that can detect a drunk passenger.

And what about misinformation, a problem already plaguing chatbots? Could bad actors incorporate misinformation into chatbots to mislead drug users about the risks?

Going back to Fritz Lang’s 1927 silent film Metropolis, audiences were fascinated by the idea of ​​a new human-like technology that makes life easier and richer. From Stanley Kubrick’s 1968 “2001: A Space Odyssey” to films like “I, Robot” and “Minority Report” from the early 2000s, however, these melancholy visions slowly morphed into something of existential dread.

It will be up to not only researchers and clinicians, but also patients and the general public, to keep AI honest and avoid turning humanity’s greatest challenges, such as the opioid epidemic, into insurmountable ones.