Professors Dessa Bergen-Cico (public health) and Asif Salekin (engineering and computer science) recruited 25 patients from Crouse Health Addiction Treatment Services in Syracuse last year to participate in their study, along with 50 others representing a healthy control group. The study measured each participant's respiration rate, heart rate, heart rate variability, skin conductance and temperature; organized the measurements into a data set; then uploaded the data onto smartwatches worn by the participants. The AI-powered watches, which were built by Salekin and two of his doctoral candidates with a grant from the National Science Foundation, also contain sensors that monitor changes in stress level, Bergen-Cico explained during a Nov. 28 interview.
In the coming months, Bergen-Cico and Salekin will monitor the participants and their smartwatches, which have been programmed to respond to elevated stress levels by instructing the patient to engage in “mindfulness practices” such as breathing exercises or meditation.
“There’s a body of research that found these practices work,” Bergen-Cico said. “The ultimate goal is to train them [patients] to use their skills and strategies.”
The measurements from the patients at Crouse confirmed what the researchers already knew: Those who suffer from opioid addiction have higher-than-normal stress responses and remain in a chronic state of anxiety, Bergen-Cico said.
The tests, she added, included interactive videos containing both relaxing and stress-inducing images, as well as a counting exercise and a game where the participants were asked to recall good and bad memories before creating a song in 30 seconds. During all of this, the research team took measurements using cameras and smartwatch sensors. None of the patients stated anything positive, yet all of them immediately recalled negative experiences, an indication of reward-deficiency syndrome, where patients are unable to recall joy or pleasure in anything, Bergen-Cico said.
In a news release from Syracuse University, Salekin said the unbiased AI algorithm he developed with his two students on the team will “accurately reflect changes in an individual’s behavior.”
“We are working with an at-risk population, so we want the AI to be robust and trustworthy,” Salekin said in the news release. “We want to avoid putting our patients in an adverse situation and incorporate that in the AI.”
If the research shows these customized smartwatches are helping patients reduce their stress levels, Syracuse University will apply for National Institutes of Health funding to cover medical trials. Bergen-Cico said she’s optimistic this treatment method could become commonplace someday.
“The medication keeps you out of the hospital, but if you don’t do anything to change your lifestyle, then you are just going day to day to stay out of trouble,” she said. “It’s really about getting to know yourself.”
