Sensor enabled wearable RFID technology for mitigating the risk of falls near beds

Abstract

The increasing ageing population around the world and the increased risk of falling among this demographic, challenges society and technology to find better ways to mitigate the occurrence of such costly and detrimental events as falls. The most common activity associated with falls is bed transfers; therefore, the most significant high risk activity. Several technological solutions exist for bed exiting detection using a variety of sensors which are attached to the body, bed or floor. However, lack of real life performance studies, technical limitations and acceptability are still key issues. In this research, we present and evaluate a novel method for mitigating the high falls risk associated with bed exits based on using an inexpensive, privacy preserving and passive sensor enabled RFID device. Our approach is based on a classification system built upon conditional random fields that requires no preprocessing of sensorial and RF metrics data extracted from an RFID platform. We evaluated our classification algorithm and the wearability of our sensor using elderly volunteers (66–86 y.o.). The results demonstrate the validity of our approach and the performance is an improvement on previous bed exit classification studies. The participants of the study also overwhelmingly agreed that the sensor was indeed wearable and presented no problems.