Carnegie Mellon University
Skip navigation and jump directly to page content

Detecting Dementia

Testing Embedded Sensors

Detecting Dementia

Carnegie Mellon University researchers in the Quality of Life Technology Center (QoLTC) will embed wireless sensors in the residences of about 50 older adults who live alone to see if they can detect subtle changes in everyday activities that indicate the onset of dementia or physical infirmities.
    
The research team is one of five nationwide selected by the Robert Wood Johnson Foundation (RWJF) to explore how observations of daily living (ODLs) — what people eat, how they sleep, their mood, how their medications makes them feel and other factors — can be captured, interpreted and integrated into clinical care.

Each team is receiving a $480,000 grant for the two-year project. The Pittsburgh team, which includes colleagues from the University of Pittsburgh Department of Rehabilitation Science and Technology and Presbyterian SeniorCare, hopes to demonstrate that simple, unobtrusive sensors in residences can alert medical professionals when a person begins to lose physical or mental abilities.
    
"The loss of the ability to make a sandwich, dial a phone, or take medications correctly often occurs gradually and, particularly for people who live alone, insidiously," said Anind Dey, associate professor in the Human-Computer Interaction Institute (HCII) of Carnegie Mellon's School of Computer Science. "If we can identify this decline at an early stage, we have a chance to halt and even reverse deterioration that might otherwise result in an unsafe living situation and ultimately require the person to be institutionalized."
    
In this initial stage of the research, Dey is working with Linda Kent, an occupational therapist at Presbyterian SeniorCare, western Pennsylvania's largest provider of care and services for older adults, to identify participants. These people will be at risk of cognitive decline, have osteoarthritis and live in one of the organization's assisted living residences. Wireless sensors will be added to such items as chairs, pillboxes and water glasses, which can then be used normally.
    
"Our goal is to determine not only that the resident has completed a task, such as preparing a meal, but how they went about it," Dey said. "Did he have trouble opening a jar? Did she take longer to make dinner than usual? Were preparation steps omitted? Over time, such changes might be a signal that a professional evaluation of a person's functional abilities is needed." The combination of motion, contact, weight and other sensors will vary from one residence to the next, depending on the occupant's needs and habits, he noted.
    
Diane Collins, assistant professor of rehabilitation science and technology at Pitt and an expert in assessing functional abilities of people with disabilities, will work with Kent to provide clinical evaluation of the participants. The other key member of the team is Matthew Lee, a Ph.D. student in HCII who specializes in behavioral and cognitive science.
    
The research effort is supported by the Robert Wood Johnson Foundation's Project HealthDesign: Rethinking the Power and Potential of Personal Health Records. Earlier Project HealthDesign work revealed that the data needed to inform day-to-day health decisions came less often from information contained in people's official medical record and more from information gained by monitoring health in everyday life.

Related Links: Read More  |  Quality of Life Technology Center  |  HCII  |  School of Computer Science


Homepage Story Archives