Smart Healthcare Bed for Prevention of Hospital Inpatient Falls

In this work we present a smart bed that has an advance warning system that can alert staff when a patient attempts to climb over the bed's side rail. Our smart bed design goes beyond traditional bed exit alarms, which simply trigger when the patient is no longer lying down. Our smart bed uses sensitive sensors and a machine-learning algorithm that identifies about-to-fall scenarios. In our tests, we achieved a prediction success rate of 96%. The system is connected to a computer screen located at the nurses' station and triggers an alarm when it predicts a fall is imminent. Multiple stations can be connected to a single data server that has an easy-to-use interface for analysis of the collected data. It can serve as a basis for a new generation of hospital smart beds.