Download files
Abstract
Research shows that older people (aged 65 and over) suffer many unintentional indoor falls, which often lead to severe injuries. As a result of an increasingly aged population in developed countries, a sizeable portion of healthcare funding is consumed in the treatment of fall-related injuries and associated long-term care. Detecting falls soon after they occur can be potentially life saving. In addition, early treatment of fall-related injuries can reduce treatment costs by minimising health deterioration resulting from long periods spent incapacitated on the floor after a fall (a scenario known as a ‘long lie’) and decreasing the number of hospital bed-days required.
In this study, an unobtrusive nighttime fall detection system based on wireless passive infrared sensors and furniture load sensors is proposed. Unlike traditional wearable sensor solutions for fall detection, which are sometimes not effective due to user compliance issues, this system causes minimal disruption to the user and alleviates privacy concerns associated with video/acoustic-based solutions.
The system was first trialed in a laboratory setting with young adults performing simulated scenarios. A preliminary algorithm was developed based on the collected data, and the hardware further refined. Later, a real-world trial was conducted at Thomas Holt retirement village in Sydney, Australia. Three subjects participated, resulting in a total of 174 days of validated data being collected. The algorithm attempted to detect two types of falls that could happen at nighttime: a fall with unconsciousness, and a fall with repeated attempted recovery. Because there were no real fall incidents reported by the subjects, false positive rates were reported, and the
final algorithm achieved on average a false alarm rate of 0.10 falls per day for falls with
unconsciousness, and 0.43 falls per day for falls with repeated attempted recovery.
This system demonstrated the feasibility of a system which performs nighttime fall detection for people living alone, with minimal invasion of privacy and a high level of user compliance. It can also be used as a complement to wearable fall detection systems to greatly improve quality of life for older people living independently.