A smart ward with a fall detection system

Fall incidents are one of the most important ward safety issues in hospitals because falls may cause extreme injuries, resulting in serious physical, psychological, and social consequences. In this paper, we present a smart ward with a fall detection system. The smart ward can issue an alarm signal to corresponding healthcare givers once it detects a fall event. The smart ward is consisted of a Kinect depth camera and a neural-network-based fall detection algorithm. We deigned six scenarios to test the performance of the proposed smart ward. The accuracy ratio was 98.15% and the KAPPA value was 0.96.