If we combine our knowledge on how people behave, why they behave and how they perceive their environments with artificial intelligence which can detect when people deviate from their normal behaviors in their customary environments, we can create "smart" environments... in this instance, potentially capable of alerting us to early stages of dementia:
"Tiny motion sensors are attached to the walls, doorways and even the refrigerator of Elaine Bloomquist’s home, tracking the seemingly healthy 86-year-old’s daily activity.
It’s like spying in the name of science — with her permission — to see if round-the-clock tracking of elderly people’s movements can provide early clues of impending Alzheimer's disease." - Associated Press, June 18, 2007.
Sensors could help catch first signs of dementia
The Alzheimer House
Wednesday, November 14, 2007
"Smart" Environments
Subscribe to:
Post Comments (Atom)
2 comments:
As far as I understand one of the first signs of AD is usually memory loss followed by deficits in executive function. These symptoms start off sub threshold. It is only when the disease progresses to visible and detectable changes in behavior that people get worried and subsequently get tested for dementia. So along these lines you would suspect that by the time there are balance and movement deficits that they would have progressed far enough into the disease that they would have already been checked.
But it is true that they need to observe older adults in their normal environment across multiple days. Their performance in neuropsych tests could greatly differ day to day based on whether it was a good day or a bad day for them. The kiosks that administer cognitive tests is an interesting idea and I wonder how effective they have been in the pilot studies. Has there been any follow up articles?
Oh, it looks like there is a paper in press.
Hayes TL, Pavel M, Kaye JA. (2007) Continuous health assessment using in-home sensors. Technology and Aging (IOS Press).
And I also found some more specifics on the research paradigm here:
http://www.bme.ogi.edu/POCL/unobtrusive_continuous.php#long
Post a Comment