Integrated Communications and Inference Systems for Continuous Coordinated Care of Older Adults in the Home
Collaborators: Misha Pavel (Northeastern University), Holly Jimison (Northeastern University), Edmund Seto (Washington University), Paul Gorman (OHSU - Oregon Health Science University)
Funding:National Science Foundation (#1111965)
Multi-sensor data collected at homes of elderly to obtain information on sleeping, physical activity, socialization etc.
Remote health care management - individual-specific models will assist in administering effective intervention to improve extend independence of elderly population.
Millions of elderly people live alone and do not take proper care of their physical health. Wireless and other sensors in home can be used to observe cognitive behavior and physical activity. The problem we wish to solve is how design, model and analysis an integrated system based on scientific foundations that are composed of several environmental and body sensors coupled with cognitive games to observe, guide and intervene healthy elderly people who live alone. The intervention will be provided by a combination of automated computer based feedback and by a human health coach. In designing the system, provisions will be made to respect the privacy of the elderly person.
In particular we are interested in observing and quantifying the levels of physical activity and model the effect of physical exercise on the cognitive levels of elderly individuals. We use cameras such as Microsoft Kinect to measure the human body kinematics and derive various features that quantify and summarize the performance of various exercises, which can either assist the coach or provide online feedback during the exercise.