Evaluation of Unconstrained Monitoring Technology Used in the Smart Bed for u-Health Environment

Abstract

Since ubiquitous technology is an emerging paradigm in healthcare, we wanted to evaluate its feasibility in the healthcare area. As a first step, we evaluated the feasibility of the smart bed. In this study, a series of experiments were conducted to evaluate the smart bed. Smart bed is a continuous ballistocardiogram monitoring device developed by Seoul National University. Eleven healthy subjects participated in the study. Each subject slept in the smart bed for two nights. To measure meaningful sleep periods, noisy sections were removed from the original signal. The subject's heart activity was obtained by automatic peak detection, using the AF2 peak detection algorithm. In this study, the feasibility of the smart bed was evaluated in terms of the meaningful sleep periods and the accuracy of automatic peak detection. On an average, 92.8% of sleep duration was meaningful. The accuracy of peak detection algorithm was also evaluated. The algorithm proved to have an accuracy of 95%; thus, we observe that 95% of the peak detection results were correct in comparison with the manual results. Based on the experimental results, we conclude that the smart bed can be a useful device for long-term monitoring in ubiquitous healthcare environments.