In order to prevent opioid users from abusing drugs, experts are working on a project to develop smart watches that can detect the emotional and psychological patterns of opioid users hours before they become addicted to drug abuse. The project was funded by the Smart Connected Health Program of the National Science Foundation, an independent agency funded by the U.S. government.
The University of Massachusetts recently announced that “a research team...has received a $1.1 million grant to further develop a smart watch sensor designed to support the long-term rehabilitation of patients with opioid use disorder (OUD).”
"More than 20 million people are struggling with substance use barriers, and the annual economic impact on the United States is 1.45 trillion US dollars in economic losses and social harm," said Yale University's Innovative Impact Program page.
According to data from the National Center for Drug Abuse Statistics, “10.1 million or 3.7% of Americans abuse opioids at least once in 12 months” and “1.6 million or 15.8% of people are eligible for opioid use disorder. The same source pointed out that in the United States, “nearly 50,000 people die from opioid overdose each year” and “for every 10 overdose deaths, at least 7 of them are caused by opioids.”
Tauhidur Rahman, an assistant professor in the School of Information and Computer Science at the University of Massachusetts and a mobile sensor expert, is collaborating with colleagues from Syracuse University and the State University of New York Northern Medical University on the project.
"We have the technology to detect these craving moments and take interventions to avoid drug use," said Rahman, who specializes in the development of health-related mobile sensors at the MOSAIC laboratory he co-led.
"This tiny wireless sensor uses machine learning to determine whether the psychophysiological signs detected in real time through breathing and electrocardiogram (ECG) are consistent with the craving for opioids," the news article explains. "This craving is one of the main reasons OUD [Opioid Use Disorder] relapses and fatal overdose after a period of abstinence."
The sensor will remind the user to pay attention to the signs of desire, and then guide the user to meditate and mindfulness-based intervention, "it can finally be personalized according to the user's behavior and clinician input."
"There is no such thing today," Rahman said. "We believe that mobile technology can provide addicts with an effective mechanism to monitor their condition and better manage their desires."
The opioid craving sensor project comes from a study published in 2019 by Rahman, lead author Bhanu Teja Gullapalli (a doctoral student working in Rahman’s laboratory) and others on the use of heart and breathing signals to sense cocaine cravings, euphoria, and drug use behavior. The press release explained.
In September, researchers published an article titled "OpiTrack: Wearable Device-based Clinical Opioids with Time Convolutional Attention Network in ACM's collection of essays on interaction, mobile, wearable and ubiquitous technology Use the tracker" thesis.
The question they asked was "Can the physiological signals obtained from wrist-worn sensors be used to detect opioid use?" 36 hospitalized subjects were used to test whether smart watches can sense and eliminate the craving for opioids.
These patients suffering from acute pain and taking opioid painkillers wear a wrist sensor that can measure their physiological signals around the clock.
The researchers wrote that patients wear "non-invasive wrist sensors (between 1-14 days) that can continuously measure physiological signals (heart rate, skin temperature, accelerometer, galvanic skin activity, and heartbeat interval)". They added that they “collected a total of 2070 hours (≈86 days) of physiological data and observed a total of 339 opioid administrations.”
The researchers used a series of mathematical processes called convolutions in the machine learning they studied.
"Once we run the convolution, we extract the features of the original data, and then train a neural network that can automatically learn to see the physical characteristics and physiological trends that indicate opioid use," Rahman explained. “So, just by looking at the watch and monitoring some parameters, we can tell when someone has taken an opioid. With our current form of technology, we have an 80% accuracy rate at a high level.”
A positive aspect of this research is that it can be applied to other substance use disorders, and the NSF-funded [National Science Foundation] project "includes machine learning courses for clinicians and awareness seminars for middle school students."
Gullapalli said the sensors in the experiment can be further used to ensure the correct use of prescribed opioid painkillers and prevent opioid use disorder (OUD).
"Doctors can require patients to wear smart watches, and the system will track how often the patient uses drugs, how the patient's physiological function changes, and determine whether the patient is dependent on opioids," Gullapalli explained, saying that the press release was recently approved by Yale University. The highly selective innovation impact plan is accepted, which is funded by the National Institute of Drug Abuse for substance use disorder entrepreneurship.
The goal of the NSF project is to promote and recognize innovative technologies. "Successful execution of the study will begin to test the effectiveness of integrating passive perception, adaptive artificial intelligence (AI) and mindfulness interventions in regulating drug cravings," the grant summary concludes.
Subscribe to our Youtube channel to get all the latest in-depth field reports from all over the world.