A framework for the elderly first aid system by integrating vision-based fall detection and BIM-based indoor rescue routing

Yuan Chen,Yuxuan Zhang,Bo Xiao,Heng Li

DOI: https://doi.org/10.1016/j.aei.2022.101766

IF: 8.8

2022-10-01

Advanced Engineering Informatics

Abstract:The occurrence of falls among older adults may result in life-threatening injuries and accidental deaths due to their vulnerability. As such, an advanced first aid system is significantly necessary to accurately detect falls and provide prompt assistance. However, current research primarily focused on fall prevention, fall detection, and first aid services after falling, thus lacking studies dealing with a systematic solution. To address this issue, the present research proposes an integrated framework for the elderly first aid system in an indoor environment using computer vision and building information model (BIM) techniques, which consists of three primary components: a vision-based module for fall detection, a cloud server (internet), and a BIM-based module for rescue routing. The experimental results showed that the proposed method could achieve 94.1% precision in identifying the fall status of older adults (i.e., falling or non-falling). Also, the proposed method enabled to automatically generate a rescue route in consideration of the routing accessibility for first aid in a BIM environment. The framework proposed in this study will improve the efficiency of the elderly first aid when falls occur, with shortening the rescue time to mitigate injury severity.

engineering, multidisciplinary,computer science, artificial intelligence