HCC: Elder-Centered Recognition Technology for the Assessment of Physical Function

点击翻译按钮获取中文摘要

Americans are living longer and more fulfilling lives. They desire to live as independently as possible. But independent lifestyles come with risks, such as debilitating falls and deteriorating health resulting from inadequate care. To address these issues, researchers are developing "smart home" technologies to enhance the safety of residents and monitor their health conditions using sensors and other devices. In particular, the continuous assessment of physical function is a key indicator of initial decline in health and functional ability. Identifying and assessing problems while they are still small can provide a window of opportunity for interventions that will alleviate the problem areas before they become catastrophic. In this project, the PI will lead a multidisciplinary team comprised of researchers in computer science and engineering, nursing, and medical informatics dedicated to developing and evaluating technology to keep older adults functioning at higher levels and living independently. They will leverage ongoing research at a unique local eldercare facility (TigerPlace) to study vision-based recognition methods for multi-person environments designed to capture continuous and automated assessments of older adults' physical function. Project objectives include: to collect video data of staged scenarios in realistic multi-person settings using older adult participants, thereby producing a body of labeled data; to utilize the collected labeled data, develop and evaluate algorithms for analyzing video in a way that preserves privacy, extracts the pose sequences of multiple persons, tracks the movement of inanimate objects, and generates assessments and summarizations of the observed activities and physical function; to evaluate the effectiveness of the summarization and assessments by showing the video and extracted information to gerontology experts and obtaining feedback; and to assess the perceptions and attitudes of older adults towards video monitoring by showing them the processed ("anonymized") video and extracted information. To achieve these goals, the PI and her team will advance the state of the art in markerless motion capture of human pose and gait, color vision-based object recognition, activity recognition in a home setting, automatic assessment of physical function through passive sensing, and understanding fundamental privacy issues for older adults.Broader Impacts: This research will impact technology, health care, policy, quality of life for older adults, and peace of mind for their families. Advances in technology have implications for other areas, including fitness and physical rehabilitation. These strides will assist health care providers to identify potential health problems and keep older adults independent longer. This, of course, means happier lives for the older adults and their families. Offering a model for eldercare technology may also provide policy makers with information to guide decisions about services for older people.