A low complexity framework for pre and post impact fall detection

Abstract

According to the studies reported by the World Health Organization newline(WHO), falls represent the second leading cause of accidental deaths around newlinethe world, producing a particularly high morbidity among people aged 65 and newlineolder (Source: worldhealth 2016). The percentage of persons residing in newlineresidential communities, with an age above 80, who experience at least one newlinefall per year is 50%, with 40% of them suffering recurrent falls. In a study of newlineinjuries due to falls among older adults (gt 60 years) in India, 65% of the newlinemembers reported prevailing injuries and 20% required hospitalizations newline(Biswas et. al. 2023). With the increase in the population of older adults newlineacross the world, the detrimental effects of falls would cause a larger strain on newlinehealthcare systems and human well-being. In terms of the economic impact newlineon the sustainability of national health systems, the global medical costs newlineattributable to falls in 2015 totaled about $50 billion. (Source: worldhealth newline2016). newlineGiven the fact that falling is a cause for concern globally across newlinedifferent environments and ethnicities, it is essential to develop a generic newlinescalable solution to detect fall. The key to developing a usable fall detection newlinesystem is to study the movement of limbs and other actions exhibited during newlinemedical conditions and then identify the type of data required to identify fall. newlineAdditionally, the components of the fall detection system such as the physical newlinesensors should be low-cost and easily available newline