Northwestern University Feinberg School of Medicine has been awarded a grant to develop wearable health sensors that prevent smoking relapse and overeating, as part of a new National Institutes of Health (NIH) Big Data initiative.
The National Center of Excellence for Mobile Sensor Data-to-Knowledge (MD2K), comprised of scientists from 12 institutions, aims to build tools that make it easier to gather, analyze and interpret health data generated by mobile and wearable sensors.
Bonnie Spring, PhD, professor in Preventive Medicine, will lead the MD2K team at Feinberg, studying Just-in-Time Adaptive Interventions to help patients quit smoking and eat healthier.
“The tools developed by the Center will lay the groundwork needed to deliver Just-in-Time Adaptive Interventions that provide people with the amount and type of help they need at the moment when they need it,” said Spring. “The studies to be conducted at Northwestern will use sensors to detect when and where someone is about to smoke a cigarette or overeat – two things that are very hard to detect as people go about their daily lives.”
After people quit smoking, they have a high risk of relapsing in the days and weeks that immediately follow. Currently, clinicians must rely on people self-reporting that they feel tempted to smoke. But often individuals feel too stressed, overwhelmed or embarrassed to do that, according to Spring.
“Now, by having the ex-smoker wear an array of sensors on a wristband and a chest band, we can mine the data to learn the patterns of movement and physiology that show when the person is smoking,” she said. “At this moment we can tell with very high accuracy from the wristband alone when the person is smoking. But we want to add more sensors that can reliably identify the 5 to 10 minutes before someone lights up. That digital pattern of signals is a signature that means that the urge to smoke is present.”
Those signals include respiration and heart rates, sweat gland activity, wrist movements and GPS.
Once clinicians can identify the precise moment of risk, they can trigger preventive interventions, such as a supportive text message or call from a coach, to deflect the temptation. The same system can also support people who are struggling to maintain a healthy diet.
“Such an intervention is just in time and can be programmed so that the algorithms adapt over time as the person’s behavior changes,” said Spring. “To prepare these capabilities for widespread use there’s much work that needs to be done to refine the usability of the sensors, the prediction algorithms and the cloud-based infrastructure that conveys the information efficiently to a panel of coaches or an electronic health record.”
Spring is director of Feinberg’s Center for Behavior and Health and co-program leader for cancer prevention at the Robert H. Lurie Comprehensive Cancer Center of Northwestern University.
In addition to Northwestern, the Center for MD2K includes scientists from Cornell Tech, Georgia Tech, Northwestern, Ohio State, Rice, UCLA, UC San Diego, UC San Francisco, the University of Massachusetts Amherst, the University of Memphis, the University of Michigan and non-profit organization Open mHealth. Across these 12 institutions the total four-year NIH grant is $10.8 million.
As a whole, the MD2K team will directly target two complex health conditions with high mortality risk – reducing hospital readmission in congestive heart failure patients and preventing relapse in abstinent smokers. The approach and products of MD2K will be applicable to other complex diseases as well, such as asthma, substance use and obesity. The Center will make the MD2K tools, software and training materials widely available and organize workshops and seminars to encourage their use by researchers and clinicians.
The Mobile Sensor Data-to-Knowledge Center is part of the NIH Big Data to Knowledge (BD2K) initiative, which is designed to support advances in research, policy and training that are needed to effectively use big data in biomedical research.