Evaluation of Three-Axial Wireless-based Accelerometer for Fall Detection Analysis

Abstract

Injuries from falling can be a serious problem for the elderly people and nowadays, a variety of research has been done on the topic of fall detection. Health professionals often refer to a person’s ability to perform Activities of Daily Life (ADL) as a measurement of their functional status. In this research, a wearable wireless-based three-axial accelerometer sensor from Shimmer has been evaluated using a falling detection algorithm from Lindemann et al. and expanded by Chia-Chi Wang et al. which is according to two parameters of sum-vector of all axes (Sa) and sum-vector of horizontal plane (Sh). These parameters are used to determine significant points of time in the falling process, and identify lying condition. For this purpose, a walking activity has been used to analyse between the fall and the ADL. According to the analysis of falls and walks through the ADL of walking activity on ten and three healthy subjects, respectively, the method is 100 percent capable to classify between the fall and the walk.