Researchers and providers alike are recognizing that human-centric smart environments can provide health monitoring services and support aging in place through adaptive interventions. The need for the development of such technologies is underscored by the aging of the population, the cost of formal health care, and the importance that individuals place on remaining independent in their own homes. The goal of this project is to design, implement, and evaluate in-home techniques for generating reports of activities and social interactions that are useful for monitoring well being and for automating intervention strategies for persons with dementia. The plan is to design machine learning techniques that make effective use of sensor data to perform automated activity monitoring and prompting-based interventions that are beneficial for the residents as well as for their caregivers and family. The environment is human-centric because it learns information about its human residents and uses this information to provide activity-aware monitoring and intervention services. By transforming everyday environments into smart environments, many older adults with cognitive and physical impairment can lead independent lives in their own homes. A key component of this project is an evaluation of the technologies in actual homes with volunteer older adults and thus will assess the technologies for acceptance with the target population.
This project addresses NSF?s Smart Health and Wellbeing goal of leveraging computational expertise leading to fundamental advances in the development of algorithms to create improvements in safe, effective, and patient-centered health and wellness services. Through design of a Gerontechnology class we are training students to design and use these technologies. This effort includes REU and IGERT students in the research project, which involves students from underrepresented groups in this multidisciplinary, collaborative effort. To facility community-wide use, comparison and collaboration, all of our datasets, tools, and course materials will be disseminated from our project web page.
