Reliability and validity analysis of MediaPipe-based measurement system for some human rehabilitation motions - LIRMM - Laboratoire d’Informatique, de Robotique et de Microélectronique de Montpellier Access content directly
Journal Articles Measurement - Journal of the International Measurement Confederation (IMEKO) Year : 2023

Reliability and validity analysis of MediaPipe-based measurement system for some human rehabilitation motions

Ameur Latreche
Ridha Kelaiaia
Adlen Kerboua

Abstract

The many epidemics that the world is witnessing necessitate the use of telehealth. Developing a website for telerehabilitation follow-up is appealing. Furthermore, the patient benefits financially from the elimination of costly equipment. The major purpose of this paper is to present a telerehabilitation website and to compare an artificial intelligence technique incorporated into the website called MediaPipe to measure range of motion with a universal goniometer and a digital angle ruler. Data was collected from around 50 healthy volunteers with the assistance of physical therapists. The MediaPipe-based shoulder measurement system's reliability is determined. To validate the approach, the 95% limits of agreement and mean difference between the MediaPipe and the two other devices are determined. The results indicate that the MediaPipe is reliable and valid for use in telerehabilitation with the studied movements. The method is then used to monitor a patient with a right patella fracture.
Fichier principal
Vignette du fichier
Article_Latreche_Final.pdf (7.02 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

lirmm-04055464 , version 1 (02-04-2023)

Identifiers

Cite

Ameur Latreche, Ridha Kelaiaia, Ahmed Chemori, Adlen Kerboua. Reliability and validity analysis of MediaPipe-based measurement system for some human rehabilitation motions. Measurement - Journal of the International Measurement Confederation (IMEKO), 2023, pp.112826. ⟨10.1016/j.measurement.2023.112826⟩. ⟨lirmm-04055464⟩
42 View
144 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More