Development of a Life-Logging Application Using Facial Expressions, Voices, and Actions for the Early Detection of Dementia and Caregiver Mental Fatigue

In this study, we will develop a life-logging application for the elderly that can record and verify behavioral patterns, facial expressions, and voice for early detection of dementia and investigate its practical application. The lifelog application displays questions, records the answers using a camera and a microphone, and saves the execution records. Since the content of the questions and the execution record are stored on the server, the administrator can view the participants’ usage status and change the questions’ content. Ten elderly people used the application for about one month, and we conducted a questionnaire survey on the usability of the application. As a result, it was confirmed that the application is usable enough, although some improvements were found, such as changing the position and size of the user interface and making questions more understandable. Based on the results, we will improve the application so that it can be used by elderly people suspected of dementia.