Development of a Nursing Care Support System that Seamlessly Monitors Both Bedside and Indoor Locations

AbstractKeeping fully abreast of a patient's daily health state is an important part of care planning and risk management. However, comprehensive assistive technology capable of collecting the data necessary for analyzing the daily health state of patients, while simultaneously monitoring other aspects their safety and security, has not yet been established. In this paper, we report on the development of two measurement systems that allow healthcare staff to more easily monitor patients within a nursing home. One is a system that measures a patient's head location on a bed (and in the bedside vicinity) via a Microsoft Kinect sensor. This system is capable of robust head location measurements via the AdaBoost-based head recognition algorithm and a random sampling consensus (RANSAC) bed recognition algorithm. The other location system tracks a patient's indoor location within the nursing home via radio frequency identification (RFID) tags. To confirm the effectiveness of these systems, we installed the proposed patient location monitoring system in an actual Tokyo nursing home and the head location tracking system in a simulated bedside environment created inside our laboratory