PIFall: A Pressure Insole-Based Fall Detection System for the Elderly Using ResNet3D

Wei Guo,Xiaoyang Liu,Chenghong Lu,Lei Jing
DOI: https://doi.org/10.3390/electronics13061066
IF: 2.9
2024-03-14
Electronics
Abstract:Falls among the elderly are a significant public health issue, resulting in about 684,000 deaths annually. Such incidents often lead to severe consequences including fractures, contusions, and cranial injuries, immensely affecting the quality of life and independence of the elderly. Existing fall detection methods using cameras and wearable sensors face challenges such as privacy concerns, blind spots in vision and being troublesome to wear. In this paper, we propose PIFall, a Pressure Insole-Based Fall Detection System for the Elderly, utilizing the ResNet3D algorithm. Initially, we design and fabricate a pair of insoles equipped with low-cost resistive films to measure plantar pressure, arranging 5×9 pressure sensors on each insole. Furthermore, we present a fall detection method that combines ResNet(2+1)D with an insole-based sensor matrix, utilizing time-series 'stress videos' derived from pressure map data as input. Lastly, we collect data on 12 different actions from five subjects, including fall risk activities specifically designed to be easily confused with actual falls. The system achieves an overall accuracy of 91% in detecting falls and 94% in identifying specific fall actions. Additionally, feedback is gathered from eight elderly individuals using a structured questionnaire to assess user experience and satisfaction with the pressure insoles.
engineering, electrical & electronic,computer science, information systems,physics, applied
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is the problem of fall detection in the elderly. Falls are a significant public health problem among the elderly, causing approximately 684,000 deaths each year. And even non - fatal falls can lead to serious health problems, such as fractures, contusions and craniocerebral injuries, which seriously affect the quality of life and independence of the elderly. Existing fall - detection methods, such as using cameras and wearable sensors, face challenges such as privacy issues, blind spots in the field of view and inconvenience in wearing. Therefore, the paper proposes a pressure - insole - based fall - detection system (PIFall), using the ResNet3D algorithm to solve these problems. Specifically, the paper designs and manufactures pressure - sensor - insole embedded with resistive films. There are 5×9 pressure sensors arranged on each insole to measure plantar pressure. These pressure data are converted into time - series "stress videos" as the input of the ResNet(2 + 1)D network for identifying falls and other actions. The paper also collects data of 12 different actions performed by five subjects, including high - risk activities specially designed to be easily confused with actual falls. The overall accuracy of the system in detecting falls is 91%, and the accuracy in identifying specific fall actions is 94%. In addition, the paper also collects feedback on the use experience and satisfaction of pressure insoles from eight elderly people through structured questionnaires to evaluate the user experience of the system. In this way, the paper aims to provide a solution that not only protects the privacy of the elderly but also enables continuous fall monitoring, while being portable and having the characteristics of low - cost maintenance.