Early detection system considering types of dementia by behavior sensing

If dementia is detected at an early stage, the progress of the disease can be slowed. Therefore, early detection of dementia is important. In this study, we propose a system for elderly people living alone which detects suspicion of dementia at an early stage of the disease. In addition, the system determines types of dementia by initial symptom of the elderly people. We compared two different analytical methods for determining initial symptom from obtained data. The first method is an analytical method by threshold and the second method is an analytical method by trend of data. We evaluated validity and immediacy of the analytical methods by scenarios.