Skip to main navigation Skip to search Skip to main content

Estimation par optimisation multi-corps de la cinématique 3D de genoux sains et arthrosiques au cours d'accroupissements : performance de modèles articulaires personnalisés

Translated title of the thesis: 3D kinematic anlaysis of healthy and osteoarthritic knee during squat with mulit-body optimisation : performance of subject-specific joint models
  • Julien Clément

Student thesis: Doctoral thesisDoctorate in Engineering: Engineering

Abstract

There is currently a lack of consensus about the movements of the knee joint. None of the methods used to analyze 3D knee kinematics is unanimously accepted by the scientific community. These methods are limited in terms of applicability in clinical practice, or in terms of soft tissue artifacts (STA) compensation. The main objective of this PhD project is to improve the functional evaluation of the knee by proposing an accurate, non-invasive and low irradiating method for measuring 3D knee kinematics. This method involves the motion capture KneeKG™ system, the multi-body optimization method (MBO), and the biplane radiographic EOS® system. First, STA of the KneeKG™ were quantified with the EOS® system during weight bearing squats performed by healthy and osteoarthritic subjects. The impact of these STA on the 3D knee kinematics were evaluated. This study shows that STA of the KneeKG™ were about 3-9° and 5-13 mm, and that they generate knee kinematic errors in the range of 9-10° and 7-10 mm. Second, the dynamic and quasi-static squats performed by the subjects were compared in terms of 3D kinematics, 3D kinetics, and electromyography of the lower limbs. This study shows that the two squatting conditions are similar. The knee kinematic differences are less than 1.5° and 1.9 mm. Third, the performance of eight combinations of joint models used in MBO to compensate for STA of the KneeKG™ were evaluated. This study shows that none of the eight combinations is ideal for correcting all the STA of the KneeKG™. Knee kinematics errors are about 13° and 7 mm after MBO. Fourth, subjectspecific knee joint models were developed from the 3D bone models derived from EOS®. These subject-specific joint models used in MBO are the most effective to correct the STA of the KneeKG™. Measurement errors are in the range of 2-6° and 2-4 mm for rotations and displacements of healthy and osteoarthritic knees. Fifth, the measurement method was used to merge the 3D kinematics and the 3D geometry of the knee, and to calculate the surface contact of the knee. Results from this study are promising. In conclusion, the combination of the KneeKG™ system, the MBO, and the EOS® system has resulted in a relatively accurate, non-invasive and low irradiating method for measuring the 3D knee kinematics during dynamic squats performed by healthy and osteoarthritic subjects.
Date23 Mar 2015
Original languageFrench
Awarding Institution
  • École de technologie supérieure
SupervisorJacques A. de Guise (Supervisor) & Nicola Hagemeister (Co-supervisor)

Cite this

'