Estimation des forces de contact fémoro-tibiales lors de la marche : comparaison de trois modèles réductionnistes

Barre, Arnaud (2007). Estimation des forces de contact fémoro-tibiales lors de la marche : comparaison de trois modèles réductionnistes. Mémoire de maîtrise électronique, Montréal, École de technologie supérieure.

Prévisualisation

PDF Télécharger (6MB) | Prévisualisation

Résumé

La détermination des forces de contact fémoro-tibiales (FCFT) nécessite l'utilisation d'une modélisation biomécanique du membre inférieur. En se focalisant sur la méthode de réduction, trois modèles estimant les FCFT ont été recensés dans la littérature. La présente étude propose de comparer ces trois modèles et de calculer leurs incertitudes pour déterminer le modèle le plus adapté à l'analyse des FCFT lors de la marche. L'estimation des FCFT est basée sur les essais de marche de sept participants asymptomatiques (27.6 (4.3) ans, 1.71 (0.10) m, 73.5 (15.0) kg), marchant à vitesse confortable (4.0 (0.4) km/h) sur un tapis roulant. Les éléments de la modélisation influençant significativement les résultats sont les bras de levier musculaires, l'ajout des gastrocnémiens et le contrôle musculaire. L'incertitude demeure pour chaque modèle inférieure à 10%. Cette étude identifie le modèle de Devita et Hortobayi (2001) comme le plus adapté à l'analyse des FCFT lors de la marche.

Titre traduit

Femoral-tibial contact forces estimation during gait: comparison of three reductionist models

Résumé traduit

The forces acting at the femoral and tibial contact surfaces are usually defined by their resultant point of force application, magnitude and orientation. Such parameters will be useful in understanding the progression of degenerative knee pathologies like osteoarthritis and prosthesis development. However, femoral-tibial contact forces (FTCF) can not be measured directly in vivo. Hence, the estimation of the FTCF necessitates the use and the development of modeling approach. In literature, severa} models have been proposed in order to estimate the FCFT. Three modeling approach, using reduction method, were developed in this study. Since no comparative study was found in literature, the purpose of the present study was twofold : to compare the three modeling approach on the estimation of the FTCF as well as to quantify their uncertainties in order to indicate which model has the greatest potential to be used in gait analysis.

This study was conducted on a group of seven healthy young adults with a mean age, height, weight of 27.6 (4.3) years, 1.71 (0.10) m, 73.5 (15.0) kg respectively. All subjects were asked to walk on treadmill at their self-paced speed. While walking in steady-state, 3D kinematics and ground reaction forces were collected in a period of 20 seconds at average velocity of 4.0 (0.4) km/h. The FCFT were estimated in the sagittal plane and averaged for 15 gait cycles. The first model (Ml) integrates two groups of muscles : the hamstring and the quadriceps muscles which were activated sequentially. This model was subdivided in three categories called respectively Mla, Mlb and Mlc. The latter differ only by their moment arm lever. The second model (M2) takes into account three groups of muscles : namely the hamstring, the quadriceps and the gastrocnemius muscle groups. Model M2 activates muscles group sequentially. Finally, the third model (M3) includes four muscle groups : the hamstring, the quadriceps, the gastrocnemius and the gluteus-maximus muscles. In model M3, the muscles are activated simultaneously.

For each model, a total of five parameters were extracted from proximal distal (PD) and anterior posterior (AP) components of the FCFT. The five parameters were compared separately using one-way analysis of variance (ANOVA) with repeated measures. The determination of model uncertainties was also quantified and analysed throughout the gait cycles. Uncertainty in the FCTF was determined on an analytical-based formulation of Taylor series of partial differential equations.

The results of this study show that there is a significant difference between models on PD and AP directions. The discrepancies between models were attributed mainly to moment arm lever, muscles group number as well as to muscular activation control. The computation of model uncertainty allows the determination of the propagation error onto the FCFT estimation. Uncertainty analysis in PD and AP directions were similar and below 10% for all models except M1b model.

This study demonstrates the influence of modelling approach onto the estimation of the FTCF, and the importance of uncertainty analysis on selection of the best model. Our data suggest that model M3 is in close agreement with recent in vivo study which measures the FTCF in a single subject with total knee prosthesis.

Type de document:	Mémoire ou thèse (Mémoire de maîtrise électronique)
Renseignements supplémentaires:	"Mémoire présenté à l'École de technologie supérieure comme exigence partielle à l'obtention de la maîtrise en génie". Bibliogr.: f. [185]-196.
Mots-clés libres:	Analyse, Calcul, Comparaison, Contact, Estimation, FCFT, Femoro-Tibial, Force, Incertitude, Marche, Modele, Reductionniste
Directeur de mémoire/thèse:	Directeur de mémoire/thèse Aissaoui, Rachid
Codirecteur:	Codirecteur Nuño, Natalia
Programme:	Maîtrise en ingénierie > Génie
Date de dépôt:	04 avr. 2011 18:13
Dernière modification:	04 nov. 2016 23:55
URI:	https://espace.etsmtl.ca/id/eprint/551

Gestion Actions (Identification requise)

Dernière vérification avant le dépôt