Design of a Smart Sole with Advanced Fall Detection Algorithm - Université Toulouse 1 Capitole Accéder directement au contenu
Article Dans Une Revue Journal of Sensor Technology Année : 2019

Design of a Smart Sole with Advanced Fall Detection Algorithm

Résumé

Fall has become the second leading cause of unintentional injury, death, after road traffic injuries, for the elderly in Europe. This proportion will increase in the next decades and become more than ever a real public health issue. France was selected by the World Health Organization to be the first country to implement a program that reduces the coverage of the dependence. Commercial automatic fall detection devices can help seniors get back on their feet faster by reducing the time of emergency procedure. Many seniors do not take advantage of this potentially life-saving technology mainly because of intrusiveness constraints. After having reminded the context and the challenges of fall detection systems, this paper presents an original device which is unobtrusive, comfortable and very effective. The hardware architecture embedded into the sole and a new fall detection algorithm based on acceleration and time thresholds are presented. The algorithm introduces a new concept of differential acceleration to eliminate some drawbacks of current systems. Tests were carried out under real life conditions by 6 young participants for different ADLs. The data were analyzed blindly. We compared the detected falls and found a 100% sensibility and more than 93% sensitivity for all participants and scenarios.
Fichier principal
Vignette du fichier
Design of a Smart Sole.pdf (882.2 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-02433657 , version 1 (28-08-2023)

Identifiants

Citer

Mostapha Zitouni, Qiang Pan, Damien Brulin, Eric Campo. Design of a Smart Sole with Advanced Fall Detection Algorithm. Journal of Sensor Technology, 2019, 09 (04), pp.71-90. ⟨10.4236/jst.2019.94007⟩. ⟨hal-02433657⟩
51 Consultations
31 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More