A threshold-based algorithm of fall detection using a wearable device with tri-axial accelerometer and gyroscope

Abstract

Fall events are the external causes of injury in the elderly adults, even leading to disability and death. In this study, we used a weightless and wearable device with built-in tri-axial accelerometer and gyroscope to record 8 types of stimulated-falls and 6 types of different activities of daily living (ADL) preformed by 6 health young subjects. A threshold-based algorithm using our device was developed to determine a fall event. Using our fall detection system, falls could be distinguished from ADL successfully for a total data set.