Elderly Fall Detection and Fall Direction Detection via Various Machine Learning Algorithms Using Wearable Sensors

Abstract

A s a result of the developments in the field of medicine, human life has been prolonged. For this reason, the ratio of elderly individuals to other individuals in the society is increasing day by day. According to the United Nations' World Population Ageing 2019 report, the world population is estimated to increase by 2 billion people and reach 9.7 billion by 2050(United Nations, 2019). According to the expectations of the World Health Organization (WHO), the number of individuals over 60 is estimated to exceed 2 billion in 2050 and the rate of deaths due to falls is estimated to increase from 28% to 42% (WHO, 2007;Chelli and Patzold, 2019). While falls of elderly people sometimes result in severe damage such as bone fracture, some falls do not cause any significant injury. The elderly person's health condition worsens as a result of the inability to get up without support after a long stay on the floor. For this reason, it is vital that the falls of elderly individuals must be detected while Article History:

Recognition of Approximate Motions of Human Based on Micro-Doppler Features

“…Non-wearable device solutions consist of vision acquisition and sensing units with image and video technologies, WiFi RF equipment and those using micro-Doppler radar, etc. [4]- [8]. The focus of our work is on human motion recognition by micro-Doppler radar sensors.…”