Hot Technologies: Technology for Independence
|Technology for independence|
An interview with Marge Skubic, PhD, Associate Professor of Electrical and Computer Engineering, University of Missouri’Columbia
|Home: It’s not only where the heart is, it’s also where the vast majority of aging Americans say they want to spend their declining years. Although high-quality skilled nursing care will always be needed, the prevailing trend in today’s long-term care policy making is toward supporting people at home-or in something resembling home, such as assisted living-for as long as possible. But there’s that nagging concern-what really is going on behind those walls? Are our aging loved ones really safe in their homes, or are they exposed to the horror of a catastrophic accident going undetected for hours or even days? Short of employing x-ray vision 24/7, what can be done to reassure families on this point? Enter the world of remote sensing technology and the work of investigators like Marge Skubic, PhD. Although remote sensing technology has already entered the long-term care marketplace (see, for example, “Where There’s Smoke…,” Nursing Homes/Long Term Care Management, April 2004), Dr. Skubic and her colleagues are carrying their investigations ever more deeply into its possibilities. They recently received a $1.2 million grant from the National Science Foundation to explore the use of a variety of sensors in unobtrusively monitoring the behavior and occasional misadventures of elderly stay-at-homes in hopes of protecting and preserving their independence. Although home care’oriented, much of their work will be conducted in a Missouri retirement community called TigerPlace, enabling her team to study some of the more institutional applications of this technology. Recently, Dr. Skubic discussed her work and its implications for long-term care in an interview with Nursing Homes/Long Term Care Management Editor-in-Chief Richard L. Peck.|
Peck: Would you explain the overall mission/goal of your research?
Dr. Skubic: We’re just getting started with it, but we have big plans. Our goals are to improve the quality of life for seniors and enable them to stay safely in their homes for as long as possible. Resident mobility and cognitive functionality are the principal focuses of our work.
Peck: How does the TigerPlace facility fit in?
Dr. Skubic: This facility affords a realistic eldercare environment. It has 32 independent living apartments, as well as common areas such as an exercise room, beauty parlor, music/activities area, and sports bar. The monitoring technology will be confined, however, to the apartments. We will use TigerPlace to develop an understanding of the needs and attitudes of seniors living in this environment. Each apartment will have at least one PC appliance (i.e., without a monitor or keyboard) connecting the sensor network with a central server to collect and process resident data.
Peck: What kind of sensors will you be using?
Dr. Skubic: There will be motion detectors; temperature monitors for the stove; a sensor mat recording pressures on the floor or selected chairs; gait monitor sensors recording, from floor vibrations, the resident’s gait style (walking, limping, shuffling, etc.), as well as falls; bed sensors recording motions of restlessness, as well as the resident’s pulse and respiration rates; and video sensors recording the resident’s daily behavior and, when they occur, accidents such as falls.
Peck: How will the video sensing work?
Dr. Skubic: We realize that this is a sensitive issue for residents and families on grounds of privacy, and I want to say at the outset that we are not recording videos for broadcast or archiving video images, and we are positioning cameras only in living room areas, not bedrooms or bathrooms. We are also using masking technology so that residents are only identified as silhouettes. Also, in some cases, we are focusing on the movement of inanimate objects rather than of the residents themselves (for example, movement of a water bottle indicating someone drinking from it). What we are really doing is mapping relative coordinates of motion against a stationary background; for example, of the water bottle moving or with a resident walking about, and we are developing algorithms to help us interpret these motions. (Interestingly, one way we are plotting indicators of a fall is to have a professor on our staff who is very good at this sort of thing act out various falls; we call him our “fall guy.”) In short, we’re doing everything we can to reassure residents and families that we are not infringing on privacy, but rather are trying to empower them to maintain independence by keeping a close eye on a resident’s activities and health status and providing useful information-and on a confidential basis, if they desire.
Peck: You mentioned the development of algorithms to track motion. How will that work?
Dr. Skubic: We are focusing on two types of activity: short term, such as taking medication, drinking fluids, and falling, and longer-term behavior defining daytime and nighttime patterns of normal behavior. Our nursing consultants from the school of nursing are helping us define what “normal” behavior might be for particular residents. Our algorithms will help us determine the likelihood of the raw sensor data being associated with activities of interest that we’re studying. We are also developing a way, using what is technically called “fuzzy logic,” to translate sensor readings into specific behavioral patterns in reports conveyed to staff and family members so that they will be able to distinguish normal behavior from activities that might be cause for concern. Behavioral “rules” are individualized and set up for each resident. For a particular resident, for example, one night’s restless sleep may not be a big deal, but if there is a pattern of restlessness, someone might need to be alerted.
Peck: Has thought been given to how this research might be adapted to more intensive care settings such as nursing homes?
Dr. Skubic: Our collaborator on this project, the University of Virginia, has investigated the use of sensor networks in some skilled facilities in Minnesota, where they tracked such issues as falls and sleeping problems. We’re adapting their sensor network and have added the video and analysis components.
Peck: What, to you, would be the “ideal outcome” of this research?
Dr. Skubic: We’d like this technology to be generally available to older people living at home. We think there is a growing recognition and demand for this approach. Recently a 79-year-old friend of mine, whose husband is 86, said we should really work hard on this so that it will be available for them when they get old!
For further information, phone (573) 882-7766 or visit www.cecs.missouri.edu/~skubic. To send your comments to the editors, please e-mail firstname.lastname@example.org. To order reprints in quantities of 100 or more, call (866) 377-6454.