LAUSR

Consumer product safety for the elderly is becoming an urgent issue globally. According to the statistics of Japan on product-related injury, the fatality rate increases with advancing age. Developing products that are safe for the elderly, including those with dementia, requires an understanding of how elderly with dementia or physical depression use consumer products. Recently an RGB-D camera with advanced artificial intelligence technology for recognizing human posture has become available. In the present study, we installed 107 RGB-D cameras (Microsoft Kinect) in five nursing homes and two ordinary homes after obtaining informed consent and captured posture data of the elderly (n=22, 70 to 102 years of age, the average age is 88 years) while they were using products such as beds, wheelchairs, wheeled walkers, doors, walking sticks, stairs, electric pots, and other electric appliances. We developed video libraries that allow us to retrieve posture data and video data from over seven hundred scenes using queries such as age, location, consumer product, degree of dementia (MMSE index), and degree of physical depression (Barthel Index). The video library software allows us to understand the relationship between the degree of dementia and consumer product use. For example, a person with severe dementia (MMSE=30%, BI=45%) could not move from a wheelchair to a bed by himself/herself, although a person with mild dementia (MMSE=56%, BI=50%) could. These examples have similar degrees of physical depression but different degrees of dementia. This comparison indicates that dementia affects the recognition function of space and the environment. Video libraries are useful for scientific understanding of the effects of dementia and physical depression on the elderly’s consumer product use. We started providing video libraries to co2870nsumer product manufacturers based on consent on appropriate use.