The underlying principles of how the nervous system selects specific muscle activation pattern, among many that produce the same movement, remain unknown. Experimental studies suggest that the nervous system may use fixed groups of muscles, referred to as muscle synergies, to produce functional motor outputs relevant to the task. In contrast, predictions from biomechanical models suggest that minimizing muscular effort may be the criteria how a muscle coordination pattern is organized for muscle synergies. However, both experimental and modeling evidence shows that stability, as well as energetic efficiency, also needs to be considered. Based on the hypothesis that the nervous system uses functionally stable muscle activation pattern for a muscle synergy, we investigated the stability of muscle patterns using a neuromechanical model of the cat hindlimb. Five unique muscle patterns that generate each of the five experimentally-identified muscle synergy force vectors at the endpoint were found using a minimum-effort criterion. We subjected the model to various perturbed conditions and evaluated functional stability of each of the five minimum-effort muscle synergies using a set of empirical criteria derived from experimental observations. Results show that minimum-effort muscle synergies can be functionally stable or unstable, suggesting that minimum-effort criterion is not always sufficient to predict physiologically relevant postural muscle synergies. Also, linearized system characteristics can robustly predict the behavior exhibited by fully dynamic and nonlinear biomechanical simulations. We conclude that functional stability, which assesses stability of a biomechanical system in a physiological context, must be considered when choosing a muscle activation pattern for a given motor task.

Postural control incorporates multiple neural and mechanical systems at various levels of the motor control system, yet the question of how all these systems interact remains unanswered. This dissertation describes development of a biologically based, three-dimensional mathematical model of the forelimb of the domestic cat that integrates skeletal anatomy, muscular architecture, and neural control. Previous work has shown that muscle architecture profoundly affects its function. However, even though the forelimbs of quadrupeds contribute to posture and locomotion differently from hindlimbs, most models of quadruped motion are based upon hindlimb mechanics. The proposed work consists of three main steps: (1) architectural and anatomical characterization, which involves acquisition of muscle attachment data, measurement of whole muscle and muscle fiber properties, and estimation of limb kinematic parameters; (2) model development and implementation, wherein the data will be integrated into a mathematical model using special-purpose software; and (3) model validation, including verification of model estimates against experimentally obtained measurements of muscle moment arms, and prediction of limb kinetics, namely end-point forces arising from perturbations to the limb. It was found that the forelimb does indeed possess structure, particularly at the shoulder and antebrachium, that allows for more diverse movements. The neural wiring in these regions is more complex than in the hindlimb, and there exists substantial muscular structure in place for non-sagittal motion and object suppression and retrieval. Other results showed that the kinematics of the limb alone produce a restorative response to postural disturbance but that the magnitude is reduced, indicating that neural input acts as a modulatory influence on top of the intrinsic mechanism of limb architecture. Furthermore the model demonstrated many of the essential features found in the experiments. This study represents the implementation of the first forelimb model of the cat incorporating mechanical properties and serves as a key component of a full quadruped model to explore posture and locomotion.

Les modèles biomécaniques éléments finis (EF) sont couramment utilisés dans de nombreux domaines. Ces modèles tendent depuis quelques années à être actifs, capable donc de générer des efforts musculaires et des mouvements. L’étape suivante consiste à rendre ces modèles réactifs, c’est-à-dire capable de réagir à une situation quelconque par des contractions musculaires et des mouvements. C’est dans cet optique que ce projet a été décomposé en 3 étapes. La première consistait à réaliser un modèle biomécanique détaillé capable de contractions musculaires et de mouvements. La seconde étape consistait à introduire des réflexes. Pour cela des modèles de capteurs sensoriels (fuseaux neuromusculaires et organes tendineux de golgi) et les réflexes associés (réflexes myotatiques et myotatiques inverses) ont ensuite été introduits au sein même du modèle. Le modèle ainsi obtenu a pu ensuite être validé grâce à une campagne expérimentale de quantification du réflexe d’étirement du tendon du biceps brachial. La dernière étape consistait à introduire des réactions de niveau supérieur, c’est-à-dire des mouvements volontaires. Pour cela une méthode de contrôle basée sur de l’apprentissage et l’optimisation a permis de générer ces mouvements et de les contrôler.En conclusion, l’introduction de boucles sensorielles et motrices de différents niveaux dans un modèle EF permet de rendre ce dernier réactif à son environnement. En effet, le modèle est ainsi capable de générer un mouvement selon des objectifs et des contraintes. Il est également capable d’adapter la contraction musculaire en fonction des évènements intervenant lors de la réalisation du mouvement. / Biomechanical finite elements (FE) models are commonly used in the field of road safety, sport and medicine. These models tended in recent years to be active, i.e. able to generate muscular efforts or movements. The next step is to make these models reactive, i.e. able to react to a situation with muscle contractions and movement. It is in this context that this project was broken down into 3 steps. The first step was to create a detailed biomechanical model capable of movements and muscle contractions. The second step was to introduce reflexes. For this, physiological sensors models (neuromuscular spindles and golgi tendon organs) and the associated reflexes associated (myotatic and inverse myotatic reflexes) were then integrated into the model. The model thus obtained could then be validated thanks to an experimental campaign of characterization of the deep tendon reflex of the biceps brachial. The last step was to introduce higher-level reactions, i.e. voluntary movements. For this purpose, a control method based on learning and optimization has made it possible to generate and control these movements.In conclusion, the introduction of sensory and motor loops of different into an FE model makes the latter reactive to its environment. Indeed, the model is thus able to generate a movement according to objectives and constraints. He is also able to adapt the muscular contraction according to the events intervening during the realization of the movement.

The main objective of the study was to investigate and simulate skeletal muscleforce production during and after isometric contractions, active muscle lengtheningand active muscle shortening. The motivation behind this work was to improve thedominant model of muscle force generation based on the theories of Hill from 1938. Effects of residual force enhancement and force depression were observed after concentric and eccentric contractions, and also during stretch-shortening cycles. It wasshown that this force modification is not related to lengthening/shortening velocity, butinstead the steady-state force after non-isometric contractions can be well describedby an initial isometric force to which a modification is added. The modification isevaluated from the mechanical work performed by and on the muscle during lengthvariations. The time constants calculated for isometric force redevelopment appearedto be in certain relations with those for initial isometric force development, an observation which extended our basis for muscle modelling. A macroscopic muscular model consisting of a contractile element, and paralleland series elastic elements was supplemented with a history component and adoptedfor mouse soleus muscle experiments. The parameters from the experiment analysis, particularly the force modification after non-isometric contractions and the timeconstants, were reproduced by the simulations. In a step towards a general implementation, the history modification was introduced in the muscluloskeletal model ofOpenSim software, which was then used for simulations of full body movements. / QC 20120525

No description available.

Les troubles musculo-squelettiques liés au travail causent des maladies physiques et mentales chez les travailleurs, réduisent leur productivité et causent de grandes pertes aux industries et à la société. Cette thèse porte sur l'évaluation du risque physique de troubles musculo-squelettiques liés au travail, pour laquelle quatre points clés sont identifiés : la mesure de la charge de travail, l'évaluation de l'effet de l'accumulation de la charge de travail, la quantification des caractéristiques individuelles et l'intégration de l'évaluation des risques dans les outils de modélisation numérique humaine. Dans l'état de l'art, les études épidémiologiques des désordres musculo-squelettiques et les méthodes actuelles utilisées pour l'évaluation des risques physiques sont présentées, ainsi que les études concernant les quatre points clés. La deuxième partie présente une étude expérimentale portant sur 17 sujets afin d'explorer un nouvel indicateur de la fatigue musculaire avec EMG de surface. Dans la partie suivante, des développements sont faits pour intégrer un modèle de fatigue musculaire dans OpenSim, un logiciel de modélisation humaine numérique, avec lequel la diminution de capacité de chaque muscle est prévisible pour une tâche donnée. Les valeurs prévues peuvent s'appliquer à l'évaluation des risques physiques. La quatrième partie présente le travail de construction d'un modèle musculosquelettique à chaîne complète dans OpenSim, étant donné qu'aucun modèle actuel ne couvre les muscles du torse et tous les membres. Une attention particulière est portée à la méthode utilisée par OpenSim pour adapter les propriétés inertielles du modèle aux individus. Les erreurs de la méthode sont évaluées à l'aide des données de référence provenant du scanner 3D du corps entier. Dans la dernière partie, le nouveau modèle de la chaîne complète est appliqué à l'analyse de la posture d'une tâche de perçage en hauteur. L'activité musculaire varie en fonction des postures, ce qui est suggéré comme indicateur des charges posturales. / Work-related musculoskeletal disorders cause physical and mental illnesses in workers, reduce their productivity and cause great losses to industries and society. This thesis focuses on the assessment of the physical risk of work-related musculoskeletal disorders in industry, for which four key points are identified: measuring workloads, assessing the effect of workload accumulation, quantifying individual characteristics and integrating the risk assessment into digital human modeling tools. In the state of the art, the epidemiologic studies of musculoskeletal disoders and the current methods used for its physical risk assessment are reviewed, as well as the studies concerning the four key points. The second part presents an experimental study involving 17 subjects to explore a new indicator to muscle fatigue with surface EMG. In the next part, efforts are made to integrate a muscle fatigue model into OpenSim, a digital human modeling software, with which the capacity decrease of each muscle is predictable for a given task. The predicted values could be applicable to the physical risk assessment. The fourth part introduce the work to build up a Fullchain musculoskeletal model in OpenSim in view that no current model covers muscles of the torso and all the limbs. Special attention is paid to the method used by OpenSim to adapt the model inertial properties to individuals. Errors of the method is evaluated with reference data coming from the whole-body 3D scan. In the last part, the newly built Full-chain model is applied on the posture analysis of an overhead drilling task. The muscle activition varies as a function of postures, which is suggested as the indicator of posture loads.

When performing a biomechanical analysis of human movement, knowledge about the ground reaction force (GRF) is necessary to compute forces and moments within joints. This is important when analysing a movement and its effect on the human body. To obtain knowledge about the GRF, the gold standard is to use force plates which directly measure all three components of the GRF (mediolateral, anteroposterior and normal). However, force plates are heavy, clunky and expensive, setting constraints on possible experimental setups, which make it desirable to exclude them and instead use a predictive method to obtain the full GRF. Several predictive methods exist. The node model is a GRF predictive method included in a musculoskeletal modeling software. The tool use motion capture and virtual actuators to predict all three GRF components. However, this model has not yet been validated during weighted leg press and weighted squat. Furthermore, the normal component of the GRF can be measured continuously during the activity with pressure sensitive insoles (PSIs), which might provide better accuracy of the GRF prediction. The objectives of this thesis were to investigate whether force plates can be exluded during weighted leg press and weighted squat and to investigate whether PSIs can improve the GRF prediction. To investigate this, the node model and a developed shear model was validated. The shear model computes the two shear GRF components based on data from PSIs, an external load acting upon the body and data from a motion capture system. Both the node model and the shear model were analysed with two test subjects performing two successive repetitions of both weighted squat and weighted leg press in a flywheel exercise device. During the leg press exercise, the node model had a mean coeffcient of correlation (Pearson's) ranging from 0.70 to 0.98 for all three directions with a mean root mean square error ranging between 8 % to 20 % of the test person's body weight. The developed shear model had a coeffcient of correlation (Pearson's) between 0.64 to 0.99 and a mean root mean square error between 3 % and 21 % of the test person's body weight. This indicates that it is possible to exclude force plates and instead predict the GRF during weighted leg press. During squat, neither the node model nor the shear model provided accurate results regarding the mediolateral and anteroposterior components of the GRF, suggesting that force plates can not yet be excluded to obtain the full GRF during weighted squat. The results of the normal component during leg press was somewhat improved with the shear model compared to the node model, indicating that using PSIs can improve the results to some extent.

No description available.

No description available.

No description available.