A Framework of Real-time Wandering Management for Person with Dementia

Recently, several solutions have been designed to detect and identify the wandering patterns for Person with Dementia (PwD). Thus, many applications have been developed which collects Spatio-temporal and contextual information from the environment. These data have not been fully utilized to its potential. The data were mainly used for the simple pattern recognition. The long-time benefit which can arise by properly managing, storing and analyzing these data has not been realized. We propose a framework for wandering data management and analytics tool which can be beneficial to the patients, caregivers as well as researchers. In this paper, we first present a novel system architecture and then discuss its key components with the integration into the system. We use web API standard for the implementation of the services. Security also forms an integral part of our design in dealing with sensitive medical data. We also demonstrate the feasibility of the system with sample analytics such as "trend in the wandering pattern" which has been done on the real-world dataset. We have done the analysis on the daytime movement data of a patient residing in Assisted Living Facility (ALF).