Design and Implementation of Bed-Exit Alarm System for Preventing Elderly Falling

In recent years, with the aging of the population, the increase in the elderly population has brought a greater burden on medical institutions. Therefore, how to reduce the incidence of hospitalization through medical technology will become an important issue for hospitals. In many hospitalization accidents, the greatest threat to the elderly due to falls. Besides immediate dangers, the sequelae will also increase the burden on caregivers and health care. In order to reduce the second-degree injury caused by falls, hospitals or nursing homes often use the bed-exit alarm system to actively inform the nurse or caregiver to help when the patient gets out of bed. However, the false alarm of related products on the market are too frequent, which not only causes the caregivers to be exhausted, but also delay critical rescue opportunities. Based on the above issues, this study will propose two new types of bed-exit alarm systems, one of which is based on the nurse's clinical care experience, setting up multiple sensing such as infrared, ultrasonic and triaxial accelerometers on the route that patients pass by most often. Another set of bed-exit alarm system uses the infrared array sensor mounted on the headboard of the bed to capture the patient's thermal imaging, and then uses the support vector machine to analyze the position of the patient on the bed to determine whether the behavior is about to leave the bed. Unlike traditional image processing, this method eliminates the need for edge detection and motion tracking and is immune to ambient light sources.