Application of multibody system techniques to human locomotor system

• Autores: Joaquín Ojeda
• Directores de la Tesis: J. Martínez-Reina (dir. tes.), Juana María Mayo Núñez (dir. tes.)
• Lectura: En la Universidad de Sevilla ( España ) en 2012
• Idioma: inglés
• Número de páginas: 195
• Títulos paralelos:
  • Aplicación de las técnicas MBS al sistema locomotor humano
• Enlaces
  • Tesis en acceso abierto en: Idus
• Resumen

  • This work can be divided in two main parts. The first one is devoted to the analysis of the motion during human gait. A study of the kinematics and kinetics of the human body during walking has been carried out and the force-shared problem has been solved with different approaches. The second one is focused on the bone remodelling analysis of a femur during normal gait. The density distribution in remodelling equilibrium with the loads produced during gait has been analyzed.

    Chapter two reviews the functional anatomy of the human body focusing on gait. The different bones, muscles and joints implied in walking are described as well as the function they play. Special interest is dedicated to muscles in order to define the way the muscle works. Besides, a detailed description of the gait cycle is introduced defining the parameters to characterize it. Finally, typical data of muscle activity during a gait cycle is provided as well as a detailed description of the role of each muscle implied in the gait.

    Chapter 3 presents the different models implemented in this work in order to model the human body and the muscles. Thus, a brief overview of multibody system techniques is shown as well as a description of the formulation employed in the work. On the other hand, the mechanical model to simulate the behavior of the muscle is presented as well as the mathematical relations used to simulate the activation and contractions dynamics of muscles. Finally, optimization algorithms employed to estimate muscle forces are presented and the cost functions used in this work are discussed.

    In chapter 4 a detailed description of the protocols used in the laboratory to carry out the measurements is presented. The different equipments used in the measurements and their characteristics are defined. Different protocols to place markers and to define joint centers which have been studied in this thesis are presented. The body pose reconstruction approach used in this work is described in detail.

    Chapter 5 describes the procedures to deal with all the measured data collected in the laboratory. Since three types of data are measured: EMG, ground reaction forces and markers trajectories, the three approaches to process the signals are explained, giving especial emphasis to the processing of the markers trajectories. As it is mentioned above, the assumption of rigid body can be violated due to relative displacements between markers. These displacements can be due to skin motions, experimental errors, etc. Different approaches to deal with this source of errors are studied and a procedure to choose the best one is proposed.

    In chapter 6 muscle forces results are presented. In this chapter muscle forces using the different algorithms proposed in chapter 3 are calculated. Besides, the influence of the different aspects of the mechanical model used to simulate the contraction dynamics has been studied. A study of the influence of the cost function defined in the optimization problem to estimate muscle forces has been developed. Results have been validated with electromyogram records and data from the literature.

    Chapter 7 shows an application of all the results calculated previously. Reactions at the joints calculated by solving the inverse dynamic problem as well as muscle forces estimated by solving an optimization problem are used as input data to estimate the density distribution on a femur. The way to do it consists in applying a bone remodelling model to a finite element model of a femur. The results are validated with the grey level extracted from CT scans. One of the main characteristics of this simulation is that all the data: loads, geometry and CT data, belong to the same subject. The bone remodelling model proposed in this thesis is a variation of a previous one, found in the literature, modified to consider cyclic loading and more particularly, the dynamical behavior of loads during gait cycle.

    Finally, in the chapter 8 a summary of the work is given together with the main conclusions and the original contributions developed in this work.