An Evolutionary Approach to Detecting Elderly Fall in Telemedicine

Fu-xing Song,Zheng-jiang Zhang,Feng Gao,Wen-yu Zhang

DOI: https://doi.org/10.1109/ccitsa.2015.26

2015-01-01

Abstract:Fall detection of the elderly is a major public health problem. Thus it has generated a wide range of applied research and prompted the development of telemedicine systems to enable the early diagnosis of fall conditions. This paper proposed a model that uses tri-axial acceleration sensor devices to detect an accidental fall and transmit the fall information to designed servers through wireless transmission devices. The fall detection algorithm we proposed is the core of this model which can be used directly in the telemedicine field. The algorithm combines Sum Vector Magnitude (SVM) and Activity Signal Magnitude Area (ASMA) to analyze the acceleration data and integrate the theory of perceptually important points (PIPs) for further analysis and judgment. The experimental result proves that our study reduces both false positives and false negatives, while improving fall detection accuracy. In addition, our solution features low computational cost and real-time response.