Rozprawy doktorskie na temat „Human gait model”

Human activity analysis has attracted great interest from computer vision researchers due to its promising applications in many areas such as automated visual surveillance, computer-human interactions, and motion-based identification and diagnosis. This dissertation presents work in two areas: general human activity recognition from video, and human activity analysis for the purpose of identifying pathological gait from both 3D captured data and from video. Even though the research in human activity recognition has been going on for many years, still there are many issues that need more research. This includes the effective representation and modeling of human activities and the segmentation of sequences of continuous activities. In this thesis we present an algorithm that combines shape and motion features to represent human activities. In order to handle the activity recognition from any viewing angle we quantize the viewing direction and build a set of Hidden Markov Models (HMMs), where each model represents the activity from a given view. Finally, a voting based algorithm is used to segment and recognize a sequence of human activities from video. Our method of representing activities has good attributes and is suitable for both low resolution and high resolution video. The voting based algorithm performs the segmentation and recognition simultaneously. Experiments on two sets of video clips of different activities show that our method is effective. Our work on identifying pathological gait is based on the assumption of gait symmetry. Previous work on gait analysis measures the symmetry of gait based on Ground Reaction Force data, stance time, swing time or step length. Since the trajectories of the body parts contain information about the whole body movement, we measure the symmetry of the gait based on the trajectories of the body parts. Two algorithms, which can work with different data sources, are presented. The first algorithm works on 3D motion-captured data and the second works on video data. Both algorithms use support vector machine (SVM) for classification. Each of the two methods has three steps: the first step is data preparation, i.e., obtaining the trajectories of the body parts; the second step is gait representation based on a measure of gait symmetry; and the last step is SVM based classification. For 3D motion-captured data, a set of features based on Discrete Fourier Transform (DFT) is used to represent the gait. We demonstrate the accuracy of the classification by a set of experiments that shows that the method for 3D motion-captured data is highly effective. For video data, a model based tracking algorithm for human body parts is developed for preparing the data. Then, a symmetry measure that works on the sequence of 2D data, i.e. sequence of video frames, is derived to represent the gait. We performed experiments on both 2D projected data and real video data to examine this algorithm. The experimental results on 2D projected data showed that the presented algorithm is promising for identifying pathological gait from video. The experimental results on the real video data are not good as the results on 2D projected data. We believe that better results could be obtained if the accuracy of the tracking algorithm is improved.

Although multiple advanced tools and methods are available for gait analysis, the gait and its related disorders are usually assessed by visual inspection in the clinical environment. This thesis aims to introduce a gait analysis system that provides an objective method for gait evaluation in clinics and overcomes the limitations of the current gait analysis systems. Early identification of foot drop, a common gait disorder, would become possible using the proposed methodology.

Osteoarthritis (OA) is a debilitating disease affecting roughly 31 million Americans. The incidence of OA is significantly higher for persons who have suffered a transtibial amputation. Abnormal cartilage stress can cause higher OA risk, however it is unknown if there is a connection between exercise type and cartilage stress. To help answer this, a tibiofemoral FEA model was created. Utilizing linear elastic isotropic materials and non-linear springs, the model was validated to experimental cadaveric data. In a previous study, 6 control and 6 amputee subjects underwent gait and cycling experiments. The resultant knee loads were analyzed to find the maximum compressive load and the respective shear forces and rotation moments for each trial, which were then applied to the model. Maximum tibial contact stress values were extracted for both the medial and lateral compartments. Only exercise choice in the lateral compartment was found to be a significant interaction (p<0.0001). No other interactions in either compartment were significant. This suggests that cycling reduces the risk for lateral OA regardless of amputation status and medial OA risk is unaffected. This study also developed a process for creating subject-specific FEA models.

The objective of this thesis is to develop and validate a computational framework based on mathematical models for the motion prediction and dynamic stability quantification of human walking, which can differentiate the dynamic stability of human walking with different mechanical properties of the leg. Firstly, a large measurement database of human walking motion was created. It contains walking measurement data of 8 subjects on 3 self-selected walking speeds, which 10 trials were recorded at each walking speed. The motion of whole-body centre of mass and the leg were calculated from the kinetic-kinematic measurement data. The fundamentals of leg property have been presented, and the parameters of leg property were extracted from the measurement data of human walking where the effects of walking speed and condition of foot-ground contact were investigated. Three different leg property definitions comprising linear axial elastic leg property, nonlinear axial elastic leg property and linear axial-tangential elastic leg property were used to extracted leg property parameters. The concept of posture-dependent leg property has been proposed, and the leg property parameters were extracted from the measurement data of human walking motion where the effects of walking speed and condition of foot-ground contact were also investigated. The compliant leg model with axial elastic property (CAE) was used for the dynamic stability analysis of human walking with linear and nonlinear axial elastic leg property. The compliant leg model with axial and tangential elastic property (CATE) was used for that with linear axial-tangential elastic leg property. The posture - dependent elastic leg model (PDE) was used for that with posture-dependent leg property. It was found that, with linear axial elastic leg property, the global stability of human walking improves with the bigger touchdown contact angle. The average leg property obtained from the measurement data of all participants allows the maximum global stability of human walking. With nonlinear axial elastic leg property, the global stability decreases with the stronger nonlinearity of leg stiffness. The incorporation of the tangential elasticity improves the global stability and shifts the stable walking velocity close to that of human walking at self-selected low speed (1.1-1.25 m/s).By the PDE model, the human walking motions were better predicted than by the CATE model. The effective range of walking prediction was enlarged to 1.12 – 1.8 m/s. However, represented by PDE model, only 1-2 walking steps can be achieved. In addition, the profiles of mechanical energies represented by the PDE model are different from that of the orbital stable walking represented by CATE model. Finally, the minimal requirements of the human walking measurements and the flexibility of simple walking models with deliberate leg property definitions allow the computational framework to be applicable in the dynamic stability analysis of the walking motion with a wide variety of mechanical property of the leg.

The extraction and analysis of human gait characteristics using image sequences are an important area of research. Recently, the focus of this research area has turned to computer vision as an unobtrusive way to analyze human motions. The applications for such a system are wide ranging in many disciplines. For example, it has been shown that visual systems can be used to identify people by their gait, estimate a subjectâ s kinematic configuration and identify abnormal motion. The focus of this thesis is a system that accurately classifies observed motions without the use of an explicit spatial or temporal model. The visual detection of hidden loads through passive visual analysis of gait is presented as a test of the system. The major contributions of this thesis are in two areas. The first is a neural network based scheme that classifies walking styles based on simple image metrics obtained from a single, monocular gray scale image sequence. The powerful neural network classifier utilized in this system provides an efficient, robust and highly accurate classification using these image metrics. This eliminates the need for more complex and difficult to obtain measures that are required by many of the currently human visual analysis systems. This system uses computer vision and pattern recognition techniques combined with physiological knowledge of human gait to estimate an observed subjectâ s hip angle. The hip angle is then used to calculate a normality index of the gait. The hip angle estimate and normality index are then used as inputs to a neural network. It is shown through experiment that this system provides an accurate classification of four different walking styles observed by a single camera. Secondly, a computer vision based approach is presented that provides an accurate pose estimate without the use of an explicit spatial or temporal model. A hybrid fuzzy neural network is used to assign contour points of a silhouette to kinematically relevant groups. These labeled points are used to estimate the joint locations of the subject. The joint angles are shown to be good estimates as compared to ground truth angles provided by a motion capture system. The effectiveness of the system to distinguish between subtle gait differences is demonstrated by detecting the presence of hidden loads when carried by walking people.
Ph. D.

Thesis: Ph. D., Massachusetts Institute of Technology, School of Architecture and Planning, Program in Media Arts and Sciences, 2018.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 111-117).
During the gait cycle, the human ankle complex serves as a primary power generator while simultaneously stabilizing the entire limb. These actions are controlled by an intricate interplay of several lower leg muscles that cannot be fully uncovered using experimental methods alone. A combination of experiments and mathematical modeling may be used to estimate aspects of neuromusculoskeletal functions that control human gait. In this research, a three-dimensional neuromuscular model of the human ankle-foot complex based on biplanar fluoroscopy gait analysis is presented. Biplanar fluoroscopy (BiFlo) enables three-dimensional bone kinematics analysis using x-ray videos and bone geometry from segmented CT. Hindered by a small capture volume relative to traditional optical motion capture (MOCAP), BiFlo applications to human movement are generally limited to single-joint motions with constrained range. Here, a hybrid procedure is developed for multi-joint gait analysis using BiFlo and MOCAP in tandem. MOCAP effectively extends BiFlo's field-of-view. Subjects walked at a self-selected pace along a level walkway while BiFlo, MOCAP, and ground reaction forces were collected. A novel methodology was developed to register separate BiFlo measurements of the knee and ankle-foot complex. Kinematic analysis of bones surrounding the knee, ankle, and foot was performed. Kinematics obtained using this technique were compared to those calculated using only MOCAP during stance phase. Results show that this hybrid protocol effectively measures knee and ankle kinematics in all three body planes. Additionally, sagittal plane kinematics for select foot bone segments (proximal phalanges, metatarsals, and midfoot) was realized. The proposed procedure offers a novel approach to human gait analysis that eliminates errors originated by soft tissue artifacts, and is especially useful for ankle joint analysis, whose complexities are often simplified in MOCAP studies. Outcomes of the BiFlo walking experiments helped guide the development of a three-dimensional neuromuscular model of the human ankle-foot complex. Driven by kinematics, kinetics, and electromyography (EMG), the model seeks to solve the redundancy problem, individual muscle-tendon contributions to net joint torque, in ankle and subtalar joint actuation during overground gait. Kinematics and kinetics from BiFlo walking trials enable estimations of muscle-tendon lengths, moment arms, and joint torques. EMG yields estimates of muscle activation. Using each of these as inputs, an optimization approach was employed to calculate sets of morphological parameters that simultaneously maximize the neuromuscular model's metabolic efficiency and fit to experimental joint torques. This approach is based on the hypothesis that the muscle-tendon morphology of the human leg has evolved to maximize metabolic efficiency of walking at self-selected speed. Optimal morphological parameter sets produce estimates of force contributions and states for individual muscles. This research lends insight into the possible roles of individual muscle-tendons in the leg that lead to efficient gait.
by David Allen Hill.
Ph. D.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2007.
Includes bibliographical references (leaf 16).
This research continues the work begun by Sungho Jo and Steve G. Massaquoi on modeling human walking and upright balance. The model of human neurological control of balance and gait generation put forward by Jo and Massaquoi in "A model ofcerebrocerebello-spoinomuscular interaction in the sagittal control of human walking" and executed in MATLAB Simulink/SimMechanics. This model has been used to determine the feed-forward command sequences for the generation of walking and running gaits. Furthermore, two feedback circuits controlling the center of mass relative to the swing leg and the composted leg angle of the simulated model were added. These provide a basis for a wider control of disturbances in order to implement running. This work helps forward the long-term goals of the MIT Sensorimotor Control Group--creating a control model of the neurological circuitry responsible for governing human balance and locomotion and testing that model by using it to control a bipedal robot. The results of this research help to prove the validity of the cerebrocerebello-spinomuscular control model developed by Jo and Massaquoi and point positively towards the introduction of the running of the control model on a physical robot.
by Ellen Cappo.
S.B.

Articular cartilage degeneration, called osteoarthritis, in the hip joint is a serious condition that affects millions of individuals yearly, with limited clinical solutions available to prevent or slow progression of damage. Additionally, the effects of high-risk factors (e.g. obesity, soft and hard tissue injuries, abnormal joint alignment, amputations) on the progression of osteoarthritis are not fully understood. Therefore, the objective of this thesis is to generate a finite element model for predicting osteochondral tissue stress and strain in the human hip joint during gait, with a future goal of using this model in clinically relevant studies aimed at prevention, treatment, and rehabilitation of OC injuries. A subject specific finite element model (FEM) was developed from computerized tomography images, using rigid bones and linear elastic isotropic material properties for cartilage as a first step in model development. Peak contact pressures of 8.0 to 10.6 MPa and contact areas of 576 to 1010 mm2 were predicted by this FEM during the stance phase of gait. This model was validated with in vitro measurements and found to be in good agreement with experimentally measured contact pressures, and fair agreement with measured contact areas.

Osteoarthritis (OA) is a degenerative condition of cartilage and is the leading cost of disability in the United States. Motion analysis experiments in combination with knee-joint finite element (FE) analysis may be used to identify exercises that maintain knee-joint osteochondral (OC) loading at safe levels for patients at high-risk for knee OA, individuals with modest OC defects, or patients rehabilitating after surgical interventions. Therefore, a detailed total knee-joint FE model was developed by modifying open-source knee-joint geometries in order to predict OC tissue stress and strain during the stance phase of gait. The model was partially validated for predicting the timing and locations of maximum contact parameters (contact pressure, contact area, and principal Green-Lagrangian strain), but over-estimated contact parameters compared with both published in vivo studies and other FE analyses of the stance phase of gait. This suggests that the model geometry and kinematic boundary conditions utilized in this FE model are appropriate, but limitations in the material properties used, as well as potentially the loading boundary conditions represent primary areas for improvement.

A marcha humana, ou caminhada, é um padrão cíclico de movimentos corporais que se repetem a cada passo que desloca um indivíduo de um local a outro. Atualmente, avaliações biomecânicas da marcha humana tem sido utilizado no diagnóstico de alterações neuromusculares, músculo-esqueléticas e como forma de avaliação pré e pós-tratamento cirúrgico, medicamentoso e/ou fisioterapêutico. O presente trabalho apresenta o desenvolvimento de uma ferramenta acadêmica de baixo custo para o estudo da marcha humana. Esse sistema consiste no sensoriamento da marcha de um usuário através de sensores inerciais e de um modelo virtual do corpo humano para permitir a visualização do movimento gerado. Dessa maneira o usuário poderá ter suas ações corrigidas por sua percepção visual e também corrigida pelas orientações de um fisiatra ou fisioterapeuta que terá a reprodução do modelo virtual conforme a movimentação detalhada do paciente para análise. O sistema ainda efetuará os registros das variáveis cinemáticas da marcha (tais como aceleração, velocidade angular, angulações dos membros sensoriados) para estudos e acompanhamento mais detalhado da sua recuperação e/ou tratamento. Como resultado, o sistema desenvolvido obteve erros médios de X 0,52º Y 1,20º Z 1,80º e erros em RMS de X 3,01º Y 3,30º Z 5,70º quando comparados com um sistema comercial, sendo esse resultado próximo à literatura e aplicável em exames biomecânicos de marcha.
The human gait is a cyclical pattern of body movements that are repeated every step that moves a subject from one location to another. Currently, biomechanical assessments of human gait has been used for diagnosing neuromuscular disorders, musculoskeletal and as a way of pre and post-surgical treatment, medication and/or physical therapy. This paper presents the development of a low cost academic tool for the study of human gait. This system consists of sensing the motion of a user through inertial sensors and a virtual model of the human body to allow the visualization of the generated movement. In this way, the user can have its actions corrected by his visual perception and also corrected by therapist or physiotherapist who will visualize the virtual model as the detailed movements of patient. The system will also record the kinematic gait variables (as acceleration, angular velocity, angles of the sensed members) for studies and more detailed monitoring of their recovery and/or treatment. As result, the developed system obtained average errors of X 0,52º Y 1,20º Z 1,80º and errors in RMS X 3,01º Y 3,30º Z 5,70º compared to a commercial system, and these results close to the ones seen in literature and applicable in biomechanical tests of gait.

Osteoarthritis (OA) is a degenerative condition of articular cartilage that affects more than 25 million people in the US. Joint injuries, like anterior cruciate ligament (ACL) tears, can lead to OA due to a change in articular cartilage loading. Gait analysis combined with knee joint finite element modeling (FEM) has been used to predict the articular cartilage loading. To predict the change of articular cartilage loading during gait due to various ACL injuries, a tibiofemoral FEM was developed from magnetic resonance images (MRIs) of a 33 year male, with no prior history of knee injuries. The FEM was validated for maximum contact pressure and anterior tibial translation using cadaver knee studies. The FEM was used to model gait of knees with an intact ACL, anteromedial (AM) bundle injury, posterolateral (PL) bundle injury, complete ACL injury, AM deficiency, PL deficiency, complete ACL rupture, as well as a bone-patellar tendon-bone (BPTB) graft. Generally, the predicted maximum contact pressure and contact area increased for all the ACL injuries when compared to intact ACLs. While an increase in maximum contact pressure and contact area is an indication of an increased risk of the development of OA, the percent of increase was typically small suggesting that walking is a safe activity for individuals with ACL injuries.

Amputation of the lower limb may result in musculo-skeletal changes similar to those that occur following space flight, immobilisation and prolonged bed rest. The similarities desist when one considers the invasive nature of amputation surgery and the impact that partial loss of a limb has on the loading characteristics of the affected leg. The aim of this study was to determine the musculo-skeletal changes that occur following trans-femoral and trans-tibial amputation, and to compare differences in the musculo-skeletal characteristics of these groups, which may occur as a function of the modified loading environment. Unique to this investigation was the study of a new trans-femoral amputee, which was incorporated to investigate the time course of any changes in muscle and bone atrophy and decreases in muscle strength in the early post-operative period. This study was also designed to provide a comparison with longer-term amputees and examine relationships between muscle morphology and strength and identified changes in gait behaviour by reference to normal gait patterns. Eight unilateral trans-femoral and 8 trans-tibial amputees (mean age 35.2yrs. ± 9.8 and 35.3yrs. ± 8.9 respectively) were subjects in the study. There were 7 males and 1 female in each of the amputee groups. A control group of similar number was used, with subjects matched on age, weight, height and gender. In the first phase of the investigation dual energy x-ray absorptiometry and magnetic resonance imaging was used to measure bone mineral density (B:MD) of the lumbar spine (L2-IA) and femoral neck (FN) and to calculate the volume and cross-sectional area (CSA) of selected muscles. Strength evaluation was assessed by measurement of maximal isometric hip torque using a Kin-Com dynamometer. Gait analyses were undertaken to determine differences in the angular kinematics of the residual and sound limb together with an electromyographic (EMG) assessment of the onset and offset of the activity of 4 hip muscles of the residual and sound thigh, which was synchronised with the kinematic measures. Differences in ground reaction force (GRF) between the residual and sound limb of each group were also examined. A single case study involving a 19-year-old motor accident victim who sustained a traumatic trans-femoral amputation of his right leg was conducted to determine the structural and functional changes over a 9-month period. Structural and functional evaluations were repeated every 3 months, beginning at 4 months post amputation, using similar methodologies and procedures described for the longer-term amputees. In the longer-term amputees the volume and CSA of the residual musculature of the trans-femoral group was significantly lower by comparison with the sound limb and no difference was found between the residual and sound musculature of the trans-tibial group. Mean torque of the residual hip was lower than that of the sound hip of the trans-femoral and trans-tibial group but the difference was not significant. Bone mineral density of the residual FN was significantly lower than that of the sound FN of the trans-femoral group but there was no difference in the trans-tibial group. No differences were found between the two experimental groups and controls at the L2-1A site. Muscle volume was significantly correlated with BMD of the L2- 1A vertebrae in the residual and sound limb of the trans-tibial group. Trans-femoral and trans-tibial amputees had a significantly slower walking velocity than that of the control group. Cadence of the trans-femoral group was significantly lower than both trans-tibial and the control groups. Stride length was not significantly different between the trans-femoral and control group but was significantly lower in the trans-tibial group. Significant differences were found between the trans-femoral and control group in the range of ankle and knee motion of the sound limb and between the sound and residual ankle, knee and hip joints of the trans-femoral and trans-tibial group. Mean GRF was lower in the residual limb compared to the sound limb for both groups, although the differences were not significant. The activity of rectus femoris, biceps femoris and adductor longus in the sound limb of the trans-femoral group were generally active for a longer duration compared to the controls and the duration of activity of these same muscles differed between the residual and sound limb musculature of the trans-tibial group. In the case study subject, BMD in the residual FN was 38.4 per cent lower than the sound FN at 4 months, decreasing to 42.1 per cent at the end of the 9-month evaluation. At this time point bone loss of the case study subject was greater than the average difference between the residual and sound limb of the longer-term amputees suggesting some recovery of bone mass may be possible. Rectus femoris and biceps femoris showed greater atrophy than the intact muscles, psoas major, adductor longus and the gluteals. At 7-months post amputation, hip torque of the residual limb in all planes of movement was lower by comparison with the sound limb. There was considerable intra-group variability in the data, which reflected the heterogeneity of the groups with respect to surgical fixation procedures, types of prosthesis used and their different physical activity levels. It was shown that longerterm trans-femoral amputees experienced considerable muscle and bone atrophy of their residual limb, which was greater than that experienced by the trans-tibial group. Although loading was not measured directly the difference between the two groups of amputees perhaps reflected their altered loading environment. Isometric hip torque was not different between the residual and sound limb of the trans-femoral and trans-tibial group, an unexpected result in the trans-femoral group considering the muscle atrophy present. The morphological changes combined with the prosthetic components were likely responsible for differences in amputee gait function. In the more recent amputee, muscle and bone atrophy was most rapid in the first 4 months but the volume and CSA of rectus femoris and biceps femoris continued to decrease up to 13 months post amputation. Isometric torque of the hip flexors and extensors decreased between 7 and 10 months and stabilised by 13 months post amputation while there was no change in the torque of the hip abductors and adductors from the initial measure. The potential for recovery of BMD, muscle size and muscle strength must be considered and may be applied to the design of more effective prostheses and rehabilitation strategies aimed at improving functional outcomes.

Every movement has a goal. For reaching, the goal is to move the hand to a specific location. For locomotion, however, goals for each step cycle are unclear and veiled by the automatic nature of lower limb control. What mechanical variables does the nervous system "care" about during locomotion? Abundant evidence from the biomechanics literature suggests that the force generated on the ground, or endpoint force, is an important task variable during hopping and running. Hopping and running are called bouncing gaits for the reason that the endpoint force trajectory is like that of bouncing on a pogo stick. In this work, I captured kinematics and kinetics of human bouncing gaits, and tested whether structure in the inherent step-to-step variability is consistent with control of endpoint force. I found that joint torques covary from step to step to stabilize only peak force. When two limbs are used to generate force on the ground at the same time, individual forces of the limbs are not stabilized, but the total peak force is stabilized. Moreover, passive dynamics may be exploited during forward progression. These results suggest that the number of kinetic goals is minimal, and this simple control scheme involves goals for discrete times during the gait cycle. Uncovering biomechanical goals of locomotion provides a functional context for understanding how complex joints, muscles, and neural circuits are coordinated.

One of the most basic reaction time experiments employed by psychologists is the comparison of latencies to responses for single and redundant targets. The general effect is that participants are capable of responding faster, that is having shorter response latencies when redundant stimuli, as opposed to an individual stimulus, are presented. Interestingly, several models attempting to predict this effect, including the well known race model, have not been entirely successful. The following study evaluated redundancy gain and violations of the race model, in three experimental models: visual only, auditory only, and a visual-auditory bimodal paradigm. The results showed redundancy gain in all three paradigms, but they were only significant violations of the race model for the visual-auditory condition. Additionally, correlations between the different paradigms were explored with respect to redundancy gain and violations of the race model on an individual participant basis.

Measurement studies have linked paraspinal muscle reflexes to low back pain. However, the role of reflexes in stabilizing the spine is not clear. Previous studies enlisted biomechanical models to aid in understanding of how intrinsic stiffness stabilizes the spine. This work expands these previous studies by modeling the neuromuscular dynamic control of the spine. The presence of delay in the reflexive system limits the availability of traditional stability analyses. However it is possible to investigate how reflex delay affects stability of the spine model using methods in linear time delayed stability. Such analyses find the maximum reflex delay, i.e., the delay margin for which stability is possible. Therefore a biomechanical model of the spine was developed that used these methods for stability. The model was able to demonstrate how reflex gains and delays affect stability. It was shown that increased proportional reflex gain reduced the amount of co-contraction required for stability. However, increased reflex gain required a reduced delay margin of the system. Differential reflex gain had no effect on the amount of co-contraction required for stability. However, it was shown to increase the delay margin for small gains. As the differential reflex gain approached the magnitude of intrinsic muscle damping the trend was reversed, and increased gain caused the delay margin to approach zero. Increased intrinsic muscle damping did not affect the minimum co-contraction required for stability, but was shown to increase the delay margin in all cases. This study provided a theoretical explanation for the role of reflexes in stabilizing the spine. Results agree with the trends in the published literature regarding patients with low-back pain. Specifically, these patients demonstrate abnormally larger reflex delay. To maintain stability, atypically small reflex gain is necessary. Compensatory co-contraction is required to offset the small reflex gain. Co-contraction and instability is observed in low back pain patients. The results presented here agree with measurement studies, and should aid in the development of hypotheses for future measurement studies.
Master of Science

Computational models of the human musculoskeletal system allow researchers to investigate human biomechanics without the need for invasive methods or expensive experiments. These models can be combined with standard motion capture technology to simulate individual’s movement patterns. With relatively little data processing, the model’s joint kinematics can be calculated along with joint kinetics, thus characterising an individual’s generalised joint coordinates (i.e., joint motion) and external joint loads. Without and with the incorporation of electromyograms (EMG) acquired during these tasks, further methods can be employed to estimate muscle forces and subsequently joint contact loading (Lloyd and Besier, 2003; Pandy and Andriacchi, 2010; Pizzolato et al., 2015; Sartori et al., 2012a; Sasaki and Neptune, 2010; Saxby et al., 2016b; Sritharan et al., 2012; Winby et al., 2009). Indeed, substantial research has focused on developing methods for estimating the magnitude of the joint contact loading within the tibiofemoral joint (TFJ) during a range of locomotion tasks (Fregly et al., 2012; Gerus et al., 2013; Kim et al., 2009). Understanding typical TFJ contact loading is crucial, as the magnitude of joint contact loading has been associated with the development and progression of TFJ osteoarthritis (Andriacchi and Mundermann, 2006). Tibiofemoral joint contact loading is primarily caused by muscles, which act to compress the joint (Sasaki, 2010). Although net and grouped muscle contributions to TFJ contact loading has previously been investigated during walking gait (Pandy and Andriacchi, 2010; Sasaki and Neptune, 2010; Saxby et al., 2016b; Sritharan et al., 2012; Winby et al., 2009), other locomotion tasks such as running and sidestep cutting, herein referred to as sidestepping remain largely unexplored. Along with the loading magnitude, other loading parameters may play a vital role influencing joint health, such as the region of loading in combination with the distribution of loading (Chaudhari et al., 2008) Models previously used to estimate the magnitude of joint contact loading have typically been linearly scaled versions of a generic musculoskeletal model, e.g., “gait2392” (Delp et al., 2007) or TLEM 2.0 (Carbone et al., 2015). These models use generic bone geometries which may not reflect each individual’s anatomy, even after linear scaling (Kainz et al., 2017a). As such, these models may be inappropriate tools to accurately estimate TFJ contact loading magnitudes, as bone geometry influence muscle tendon unit (MTU) force estimates and contact mechanics (Demers et al., 2014; Gerus et al., 2013; Lerner et al., 2015). Furthermore, these models may be inappropriate for the estimation of the region of loading within the TFJ, as this feature is highly dependent on joint anatomy (Lerner et al., 2015). Additionally, limitations within these linear scaled generic models, particularly the TFJ kinematic models, further hamper their utility for investigating regional loading within the TFJ (Demers et al., 2014). If regional loading is to be investigated, computational models that accurately represent subject-specific three-dimensional (3D) bone and joint geometry, 6 degree of freedom (DOF) joint kinematics, and feasible MTU pathways, lengths, and moment arms are required. The overarching aim of this thesis was to investigate features related to TFJ contact loading. Specifically, estimate individual muscle contributions to medial and lateral TFJ contact loading during walking, running, and sidestepping. Second, develop a framework that automatically creates and tunes highly detailed subject-specific computational musculoskeletal models that can be used to investigate various feature of TFJ loading. To investigate individual muscle contributions to TFJ contact loading during various dynamic locomotion tasks, 54 healthy individuals were recruited as part of an ongoing project. Each participant underwent a standard motion capture gait analysis session, wherein whole body and segment motions were captured using 3D motion capture. Ground reaction forces were acquired via in-ground force plates and muscle activation patterns acquired via surface EMG. Motion capture data were used within the free and open-source musculoskeletal modelling platform OpenSim (Delp et al., 2007) to estimate model joint kinematics and kinetics. Using established calibrated EMG-informed neuromusculoskeletal modelling methods (Hoang et al., 2018; Pizzolato et al., 2015; Saxby et al., 2016b), muscle forces and subsequently muscle contributions to TFJ contact loading (Winby et al., 2009) were estimated. Results for walking, running, and sidestepping showed during weight acceptance, the vastus medialis and vastus lateralis muscles dominated contribution to medial and lateral TFJ contact loading respectively. During mid-stance and push-off, the contribution to medial and lateral TFJ contact loading was dominated by the medial and lateral gastrocnemii muscles respectively for all three tasks. Although there were similarities in which muscles dominated medial and lateral TFJ contact loading, differences were shown in the magnitude of relative muscle contributions between locomotion tasks. These differences were driven by different kinematic (Novacheck, 1995), kinetic (Novacheck, 1995), muscle activations (Besier et al., 2003a), and stabilisation requirements present in each tasks. Specific differences were, quadriceps contribution to medial and lateral TFJ contact loading were higher during running compared to walking, while gastrocnemii contribution to medial and lateral TFJ contact loading were higher during walking compared to running. Comparing running and sidestepping, contribution of selected muscles to medial TFJ contact loading were higher during sidestepping, while selected muscle contributions to lateral TFJ contact loading were higher in running. Muscles which dominate the contribution to TFJ contact loading, also provide a majority a TFJ stabilisation during these tasks. Results may provide valuable information for rehabilitation following orthopaedic surgeries to restore TFJ stability and prevent future injuries. To further address the overarching aims of this thesis, highly detailed subject-specific musculoskeletal models were required and thus developed. Subject-specific musculoskeletal models may contain joints and MTU pathways that are both physically and physiologically infeasible. First, the articulating bones of joints may interpenetrate, and similarly MTUs can penetrate bone surfaces. Second, joint kinematics can have discontinuities and may not follow the patterns of previously reported cadaveric studies. Likewise, MTU pathways, if inappropriately defined, can create MTU lengths and moment arms (MTU kinematics) that exhibit discontinuities and do not follow patterns of previously published cadaveric data. This thesis created subject-specific musculoskeletal models that addressed these shortcomings along with shortcomings of previous modelling methods. To evaluate the framework developed to create detailed subject-specific musculoskeletal models, a set of 6 individuals from an on-going study were used. These subjects spanned age (21 – 32 years), height (160.5 – 185 cm), and mass (45 – 89 kg) ranges, and were composed of three females and three males. Each subject underwent a standard gait analysis session as well as a comprehensive magnetic resonance imaging (MRI) protocol enabling detailed visualisation of their bones, muscles, cartilages, and other articular structures (i.e., ligaments). The framework developed within this thesis was built atop a pre-existing open-source framework, the Musculoskeletal Atlas Project (MAP) Client (Zhang et al., 2014), written in Python (Python Software Foundation. Python Language Reference, version 2.7. Available at http://www.python.org). The MAP Client was used in combination with manually segmented MRIs to well reconstruct subject-specific bone geometry using direct image segmentation and pre-developed MAP Client statistical shape models (SSMs). Bone geometries were then used to customise a generic OpenSim model (Delp et al., 2007) with subject-specific bone geometries and other personalised features (i.e., joint positions , MTU origin and insertions, and MTU via points). This generated model represents the standard model produced via the MAP Client pathway; however a number of further developments were required. Required improvements ranged from simple inclusions, such as adding customised marker sets, patellae, and patellofemoral joints (PFJs). More complex additions involved the definition of 6 DOF TFJ and PFJ kinematics, MTU origins, insertions and pathways. Six DOF (1 independent and 5 coupled) subject-specific TFJ and PFJ mechanisms were built from segmented MRIs. These mechanisms were then automatically tuned to be physically and physiologically feasible using previously published methods (Brito da Luz et al., 2017), which were incorporated into the MAP Client framework. The MTU pathways, i.e., origins, insertions and wrapping surfaces, were defined using the MAP Client mean SSMs and subject-specific MAP Client-generated bone reconstructions, which were automatically tuned to be physically and physiologically feasible (discussed in detail later). In the present study, MTU origins and insertions were defined using an atlas-based method (Zhang et al., 2015). The MTU origins and insertions were positioned on the MAP Client mean SSMs in the same anatomical regions (i.e., node point) as the atlas and could be queried using node indices. Node indices were used to define subject-specific MTU origin and insertions on the subject-specific MAP Client-generated bone models. The MTU wrapping surfaces were defined in a two-step process: (i) placement and fitting of wrapping surfaces, and (ii) optimisation of the wrapping surfaces’ geometrical dimensions and positions. Initial selection and placement of wrapping surfaces was done manually based on anatomical regions and landmarks defined as bone mesh elements and nodes indices on the MAP Client mean SSM bones. Following this manual identification, using individual’s subject-specific bones, wrapping surfaces were automatically positioned using bone elements and nodes identified through the MAP Client using custom written Python (Python Software Foundation. Python Language Reference, version 2.7. Available at http://www.python.org) software. Wrapping surface dimensions were based on analytical shapes automatically fit to anatomical regions of subject-specific MAP Client-generated bones using custom written Python software. Once wrapping surfaces were placed, their positions, orientations, and dimensions were automatically optimised with the aim of producing MTU pathways that were physically and physiologically feasible. “Physically feasible” MTU pathways (i) did not penetrate bones, and (ii) did not produce non-sensible wrapping scenarios, such as completing a circumferential loop of a wrapping cylinder. “Physiological feasible” MTU lengths and moment arms (i) closely follow the pattern of measurements taken from cadavers, available in literature, and (ii) are free of discontinuities. The developed framework created and tuned subject-specific rigid body musculoskeletal models that largely produced the desired outcomes while overcoming limitations with previous modelling methods. With respect to the definition of physically feasible MTU pathways, the inclusion of MTU wrapping surfaces fit to each subject’s anatomy largely reduced the number of bone MTU penetrations. However, prior to tuning, including MTU wrapping surfaces fit to each subject’s anatomy were detrimental to many MTU kinematic metrics (i.e., MTU kinematic smoothness, and pattern similarity to literature data). The designed optimisation routine, to tune MTU wrapping surfaces provided further improvements to MTU pathways, and more importantly improved MTU kinematic smoothness and pattern similarity with literature data. Improvements to both MTU pathways and kinematics was present in models which contained simplified (MAP Client standard) as well as subject-specific 6 DOF TFJ and PFJ kinematic mechanisms. The fact that improvements were shown in both models, regardless of the joint model, provided further confidence in the MTU wrapping surface optimisation process that was developed. Additionally, the fact that improvements were only shown once tuned, provided further evidence that the tuning of subject-specific musculoskeletal models is a necessary step in the developed framework. The optimised subject-specific musculoskeletal models containing simplified joint models, compared to subject-specific joint models, performed more consistently and often produced favourable MTU kinematics and pathways. Models with simplified joint models often exhibited fewer bone MTU penetrations, smoother MTU kinematics, and kinematics which more closely matched the pattern of previously reported cadaveric data. The less consistent results seen in models with subject-specific joint kinematics may be due to greater inter-subject variability within estimated kinematics for both the TFJ and PFJ. Further improvements to both the developed MTU wrapping surface optimisation framework and joint kinematic models may produce more consistent results for models with subject-specific joint kinematic models. Although not implemented within this thesis, the models produced using this proposed framework can be used to investigated the regional loading of the TFJ during a range of locomotion and other dynamic tasks. The framework presented here represents a large advancement of the field of subject-specific computational modelling. Along with addressing a number of short comings of previous models, the methods presented in this thesis are predominately automated which reduces the time and cost burdens which are typically associated with building subject-specific models. These time and cost burdens are related to the collection and segmentation of full lower limb MRI. Additionally, all developed methods utilise free and open-source software, facilitating the sharing and wider adoption of these subejct specific methods and models. Improvments to these developed models and methods facilitate their use in both academic research as well as a number of clinical and medical applications. The highly automated and tuned framework reduces both the time and knowledge burden on the user, providing further advantages to these methods.
Thesis (PhD Doctorate)
Doctor of Philosophy (PhD)
School Allied Health Sciences
Griffith Health
Full Text

Thesis (Ph. D.)--Mechanical Engineering, Georgia Institute of Technology, 2008.
Wayne J. Book, Committee Member ; Young-Hui Chang, Committee Member ; T. Richard Nichols, Committee Member ; Lena H. Ting, Committee Co-Chair ; Stephen P. DeWeerth, Committee Co-Chair.

Cette thèse examine la relation complexe et la causalité entre la migration et le développement, sujet d'actualité vus les flux croissants de migrants et les transferts de fonds privés. La revue de la littérature théorique et empirique montre la complexité de cette relation, et l’absence de consensus dégagé par les travaux menés sur les causes et les effets de la migration sur le développement. Sur les causes de migration, une première estimation empirique montre qu’elle fait partie intégrante du processus de développement et n’est donc pas une simple conséquence de faibles niveaux de développement: le niveau de développement des pays d'origine s’accroissant, les aspirations et les capabilités des populations augmentent et si celles-ci font face à l'absence d’opportunités, elles vont migrer à condition d’en avoir les capabilités (compétences requises, moyens financiers, politiques de migration, etc.). Concernant l’impact de la migration, une deuxième estimation empirique montre un effet positif sur le développement via les transferts privés. Les modèles indiquent que leur contribution se fait à travers deux principaux canaux, l'investissement en capital et le capital humain (éducation et santé), lesquels sont susceptibles de permettre un développement à long terme des pays d'origine. Au niveau micro, une troisième série de modèles étudie le mécanisme de cet impact au niveau de ménages, à partir du cas de l'Egypte. Ces modèles confirment l'importance des transferts privés sur les niveaux d'éducation et de santé dans les ménages qui les reçoivent. Ces résultats sont censés contribuer à la compréhension de cette relation complexe entre migration et développement
This thesis is concerned with the causal and complex relation between migration and development. A timely subject, especially with increasing flows of migrants and the remittances these migrants send home. Both the theoretical and empirical literature reviews address the complexity of this relation but consensuses on the causes and impacts of migration on development are generally inconclusive. On the causes of migration, our first empirical estimation shows that migration is part of the development process and not a simple result of its low levels: the increasing development level of the home countries increase the aspirations and capabilities of their populations and if these are faced with lack of opportunities at home, individuals seek migration provided they have the right capabilities (skills required, financial means, migration policies, etc.). On the impact of migration, our second empirical estimation gives evidence of a positive impact through remittances on the development of the home countries. The models show the positive contributions of remittances towards development through two main channels: capital investment and human capital (education and health). These two channels are believed to achieve long-term development of the home countries. At the micro level, we look at the mechanism of this impact at the household level, addressing the case study of Egypt. Our third models give evidence of the importance of these remittances in increasing both education and health status of the recipients’ households’. These findings are believed to make a contribution towards the understanding of this complex relation between migration and development

This research presents a novel approach to human gait analysis using wearable Inertial Measurement Unit(IMU) sensor-based technique.The proposed system emphasizes on detection of certain abnormal gait pattern. It includes hemiplegic and equinas gait which are synthetically generated in our lab.The designed prototype contains an IMU sensor with 3 axial accelarometer and gyroscope. It provides linear accelaration and angular velocity of human foot.A probabilistic framework, Hidden Markov Model(HMM) is applied to model bipedal human gait.This model uses Symbolic Aggregate Approximation(SAX) method for generating observation sequences obtained from sample gait cycles.The detection of abnormal gait pattern is based on maximum log-likelihood of an unknownobserverd sequence,generated from a gait cycle.The experimental results demonstarte that the proposed HMM-based technique is able to detect gait abnormality in gait data.The proposed personalized gait modelling approach is cost effective and reliable to implement in gait rehabilatation process.

Abstract The analysis of biomechanical leg is a very important clinical instrument to determine the nature of disease. Now a day many researchers analysed on biomechanics system such as human gait. The modelling and simulation of human gait is developed through various techniques such as vectorial mechanics Lagrange`s equations, Kane method. These dynamic model of human body illustrate a different positional forces which is exerted on a human leg like a muscle, tendon, and other physiological effect. One of the techniques to simulate the human gait is bond graph modelling techniques. This technique avails to develop the model in multi energy domain. The generated differential equation is further used for simulation. . The present work proposed a computational model of human gait through bond graph modeling and simulation work is carried out on Symbols Sonata© software. The fifth order Runga-Kutta method used for iteration. This work also revealed the introduction and different methodologies which used in the analysis of human leg e.g., inverse dynamic simulation and direct dynamic simulation and also explain the cyclic nature of gait. Keywords: Human gait model, Bond graph modelling, Dynamic simulation

Dissertação de mestrado integrado em Biomedical Engineering (área de especialização em Biomaterials, Biomechanics and Rehabilitation)
Ankle-foot orthoses are orthotic devices that support the ankle joint and are appropriate for several pathologies, mostly the ones that cause dropfoot, which is caused by an ankle joint deficiency. In the present work, a planar multibody model of the human body in the sagittal plane was developed. For this purpose, the MOBILE computational program was utilized. The model simulates the lower limbs and is made of 9 rigid bodies. It has 12 DOFs and is prepared for reproducing kinematic data acquired in a gait lab. Kinematic measurements were obtained in a gait lab from a healthy subject, with and without plastic ankle foot orthoses worn on both feet. The results obtained showed that with the orthoses, the ankle joint behavior is similar to a linear torsional spring, with almost no hysteresis. Ankle kinematics, measured in the gait lab with and without orthoses, were successfully reproduced by forward dynamics using the multibody model developed, which allows for the validation of the presented approach. Furthermore, it was concluded that ankle foot orthoses can be modeled as a spring element acting at the ankle joint, and the use of an ankle foot orthosis reduces the muscle activation at the ankle in about 15%.
Knöchel-Fuß-Orthesen sind orthopädische Geräte, die das Sprunggelenk zu unterstützen und sind für verschiedene Erkrankungen, vor allem diejenigen, die dropfoot, die von einer Sprunggelenks-Mangel verursacht wird dazu führen, angemessen. In der vorliegenden Arbeit wurde ein planarer Multibody Modell des menschlichen Körpers in der Sagittalebene entwickelt. Zu diesem Zweck wurde das MOBILE Rechenprogramm verwendet. Das Modell simuliert die unteren Extremitäten und wird von 9 starren Körpern. Es verfügt über 12 Freiheitsgrade und ist für die Wiedergabe kinematische Daten in einer Ganglabors erworbenen vorbereitet. Kinematische Messungen wurden in einem Ganglabors von einer gesunden Person erhalten, mit und ohne Kunststoff Sprunggelenk Orthesen an beiden Füßen getragen. Die erhaltenen Ergebnisse zeigten, dass mit den Orthesen, das Sprunggelenk Verhalten ähnlich einer linearen Torsionsfeder ist, fast ohne Hysterese. Knöchel Kinematik, in der Ganglabors mit und ohne Orthesen gemessen wurden erfolgreich von Vorwärtsdynamikanalyse Verwendung der Multibody Modell entwickelt, das für die Validierung der dargebotenen Ansatz ermöglicht reproduziert. Darüber hinaus wurde festgestellt, dass Sprunggelenk Orthesen als Feder wirkende Element am Sprunggelenk modelliert werden können, und die Verwendung eines Fußheberorthese verringert die Muskelaktivität am Knöchel in etwa 15%.
As ortóteses do tornozelo são dispositivos ortopédicos que apoiam a articulação do tornozelo e são indicados para uma variedade de patologias, nomeadamente as que causam pé pendente, que é uma deficiência na mobilidade do tornozelo. No presente trabalho, um modelo multibody planar do corpo humano no plano sagital foi criado. Para tal, o software MOBILE foi usado. O modelo simula os membros inferiores e é composto por nove corpos rígidos. Possui 12 graus de liberdade e está preparado para usar dados cinemáticos adquiridos num laboratório de análise da marcha humana como restrições de guiamento. Dados cinemáticos foram obtidos num laboratório de análise da marcha humana, a partir de um indivíduo saudável, com e sem ortóteses plásticas em ambos os pés. Os resultados mostraram que, com a ortótese, o comportamento da articulação do tornozelo é semelhante a uma mola de torção linear, praticamente sem histerese. Os dados cinemáticos do tornozelo, medidos no laboratório de marcha, com e sem ortótese, foram reproduzidos com sucesso por uma dinâmica direta, utilizando o modelo multibody desenvolvido, o que validou a abordagem utilizada. Todas as metodologias encontram-se descritas e explicadas nesta tese e concluiuse que a ortóteses do tornozelo podem ser modeladas como uma mola de torsão que actua na articulação do tornozelo. Concluiu-se também que a utilização de uma ortótese do tornozelo por uma pessoa saudável reduz a activação muscular do mesmo em cerca de 15%.

L'évaluation de la démarche humaine est l'une des composantes essentielles dans les soins de santé. Les systèmes à base de marqueurs avec plusieurs caméras sont largement utilisés pour faire cette analyse. Cependant, ces systèmes nécessitent généralement des équipements spécifiques à prix élevé et/ou des moyens de calcul intensif. Afin de réduire le coût de ces dispositifs, nous nous concentrons sur un système d'analyse de la marche qui utilise une seule caméra de profondeur. Le principe de notre travail est similaire aux systèmes multi-caméras, mais l'ensemble de caméras est remplacé par un seul capteur de profondeur et des miroirs. Chaque miroir dans notre configuration joue le rôle d'une caméra qui capture la scène sous un point de vue différent. Puisque nous n'utilisons qu'une seule caméra, il est ainsi possible d'éviter l'étape de synchronisation et également de réduire le coût de l'appareillage. Notre thèse peut être divisée en deux sections: reconstruction 3D et analyse de la marche. Le résultat de la première section est utilisé comme entrée de la seconde. Notre système pour la reconstruction 3D est constitué d'une caméra de profondeur et deux miroirs. Deux types de capteurs de profondeur, qui se distinguent sur la base du mécanisme d'estimation de profondeur, ont été utilisés dans nos travaux. Avec la technique de lumière structurée (SL) intégrée dans le capteur Kinect 1, nous effectuons la reconstruction 3D à partir des principes de l'optique géométrique. Pour augmenter le niveau des détails du modèle reconstruit en 3D, la Kinect 2 qui estime la profondeur par temps de vol (ToF), est ensuite utilisée pour l'acquisition d'images. Cependant, en raison de réflections multiples sur les miroirs, il se produit une distorsion de la profondeur dans notre système. Nous proposons donc une approche simple pour réduire cette distorsion avant d'appliquer les techniques d'optique géométrique pour reconstruire un nuage de points de l'objet 3D. Pour l'analyse de la démarche, nous proposons diverses alternatives centrées sur la normalité de la marche et la mesure de sa symétrie. Cela devrait être utile lors de traitements cliniques pour évaluer, par exemple, la récupération du patient après une intervention chirurgicale. Ces méthodes se composent d'approches avec ou sans modèle qui ont des inconvénients et avantages différents. Dans cette thèse, nous présentons 3 méthodes qui traitent directement les nuages de points reconstruits dans la section précédente. La première utilise la corrélation croisée des demi-corps gauche et droit pour évaluer la symétrie de la démarche, tandis que les deux autres methodes utilisent des autoencodeurs issus de l'apprentissage profond pour mesurer la normalité de la démarche.
The problem of assessing human gaits has received a great attention in the literature since gait analysis is one of key components in healthcare. Marker-based and multi-camera systems are widely employed to deal with this problem. However, such systems usually require specific equipments with high price and/or high computational cost. In order to reduce the cost of devices, we focus on a system of gait analysis which employs only one depth sensor. The principle of our work is similar to multi-camera systems, but the collection of cameras is replaced by one depth sensor and mirrors. Each mirror in our setup plays the role of a camera which captures the scene at a different viewpoint. Since we use only one camera, the step of synchronization can thus be avoided and the cost of devices is also reduced. Our studies can be separated into two categories: 3D reconstruction and gait analysis. The result of the former category is used as the input of the latter one. Our system for 3D reconstruction is built with a depth camera and two mirrors. Two types of depth sensor, which are distinguished based on the scheme of depth estimation, have been employed in our works. With the structured light (SL) technique integrated into the Kinect 1, we perform the 3D reconstruction based on geometrical optics. In order to increase the level of details of the 3D reconstructed model, the Kinect 2 with time-of-flight (ToF) depth measurement is used for image acquisition instead of the previous generation. However, due to multiple reflections on the mirrors, depth distortion occurs in our setup. We thus propose a simple approach for reducing such distortion before applying geometrical optics to reconstruct a point cloud of the 3D object. For the task of gait analysis, we propose various alternative approaches focusing on the problem of gait normality/symmetry measurement. They are expected to be useful for clinical treatments such as monitoring patient's recovery after surgery. These methods consist of model-free and model-based approaches that have different cons and pros. In this dissertation, we present 3 methods that directly process point clouds reconstructed from the previous work. The first one uses cross-correlation of left and right half-bodies to assess gait symmetry while the other ones employ deep auto-encoders to measure gait normality.

Walking comes naturally to us and appears to be simple. However, this is not so and it is known that walking requires high level neural control and muscle coordination. There is no single, unifying theory of bipedal walking. Models of walking are useful in various ways such as developing computational theories of neural control, understanding muscle coordination and to design and analyze lower extremity prostheses. This thesis deals with modeling and simulation of walking from a kinematics and dynamics view point. Three sagittal planar models with increasing levels of complexity are presented in this thesis. The first model is a simple two degrees of freedom (DoF) model representing the motion at the hip and the knee joint. The second model is a three DoF model where the ankle joint motion is also taken into account. Finally, the third model considers both the legs and has seven DoF. The kinematic and dynamic equations of the models are derived, and the inverse dynamic analysis and forward dynamic simulation of the models are performed. The simulation results are compared with experimental data available in literature.

Walking comes naturally to us and appears to be simple. However, this is not so and it is known that walking requires high level neural control and muscle coordination. There is no single, unifying theory of bipedal walking. Models of walking are useful in various ways such as developing computational theories of neural control, understanding muscle coordination and to design and analyze lower extremity prostheses. This thesis deals with modeling and simulation of walking from a kinematics and dynamics view point. Three sagittal planar models with increasing levels of complexity are presented in this thesis. The first model is a simple two degrees of freedom (DoF) model representing the motion at the hip and the knee joint. The second model is a three DoF model where the ankle joint motion is also taken into account. Finally, the third model considers both the legs and has seven DoF. The kinematic and dynamic equations of the models are derived, and the inverse dynamic analysis and forward dynamic simulation of the models are performed. The simulation results are compared with experimental data available in literature.

Due to the ongoing trend of an ageing population in many developed countries, falls among older adults during walking are emerging as an important social healthcare issue due to high injury rates and associated medical costs. Unsuccessful recovery from balance loss leads to falls, and four critical swing phase gait events are toe-off, heel contact, minimum foot clearance (MFC) and minimum lateral margin (MLM). Dynamic balance at these four gait events were examined to characterise biomechanical evidence in older adults’ walking patterns to understand the frequent occurrence of falls. Falls prevention strategies need to be effective and practical in addition to cost advantages and ease of engagement, if a measure is to be adopted as a long-term habitual intervention. This project also investigated whether a shoe insole is effective in reducing falls risks. As insoles can be applied to any shoes at lower cost, insole interventions could possibly prove an ideal approach to falls prevention during walking.

Currently normalcy indices have not been used to evaluate the gait of normal healthy adults. The purpose of this study was to assess the appropriateness of using a normalcy index in assessing gait patterns by investigating normalcy indices of different compositions created from an adult population. Thirty-seven normal and healthy subjects between the ages of nineteen and sixty-six underwent a clinical gait analysis. Ten trials from the left and right sides of the body were averaged. From the averaged data, sixteen kinematic variables were chosen for further analysis. Principal component analysis was applied to the variables and four separate normalcy indices were formed. They consisted of a right side, left side, left and right combined, and average of both sides indices. For the left side index, 23.41% of the variance was accounted for in the principle component. In the right side index, 25.35% of the variance was accounted for. 25.45% and 23.95% of the variance was accounted by the PC in the average of both sides and the left and right combined indices. The mean index scores of the left and right indices are 6.48 and 8.38. Index score for the average of both sides and left-right combined indices is 6.80. A paired samples t-test showed that the left and right indices were significantly different at the .01 level. The mean of the index from both sides and the left-right combined were not significantly different with an alpha of 0.969. The significant difference between the left and right side indices suggests that with further refinement, normalcy indices may possibly be used to assess the symmetry of gait.
School of Physical Education, Sport, and Exercise Science

Thesis (Ph. D.)--Florida State University, 2005.
Advisor: Anuj Srivastava, Florida State University, College of Arts and Sciences, Dept. of Statistics. Title and description from dissertation home page (viewed Jan. 24, 2006). Document formatted into pages; contains xvii, 158 pages. Includes bibliographical references.

碩士
國立成功大學
醫學工程學系
84
Recently, scholars have used biomechanical model to predict and analyze the dynamics of human motion. Some of them focus on gait analyses that intend to give people better understanding of human walking pattern and the induced joint forces. In order to build a biomechanical model that can predict the gait motion of below-knee amputees, this paper uses seven links to build a three dimensional dynamic model to simulate the swing phase of human gait. The dynamic equations are derived based on Newton- Euler method. Once the motion parameters such as displacement, velocity、and acceleration are measured experimentally,the joint forces of normal subject could be obtained by inverse- dynamic method. These joint forces could then be used as referenced parameters to predict the motion trajectory of below- knee amputees by direct-dynamic method, where some of the parameters, such as the body mass and moment of inertia, are changed to that of amputees, also some constraints are added to limit the mobility of the model to emulate an artificial limb .

Doutoramento em Motricidade Humana na especialidade de Biomecânica
A gravidez é uma fase especial da vida , considerando as adaptações morfológicas, fisiológicas, biomecânicas e hormonais vivenciadas pelas mulheres durante cerca de 40 semanas e no período pós-parto, podendo modificar o padrão de marcha e contribuir para uma sobrecarga no sistema músculo-esquelético, causando dor nos membros inferiores, bacia e zona lombar. Os objetivos do presente trabalho foram: 1) analisar a marcha de mulheres grávidas no segundo trimestre; 2) comparar as adaptações biomecânicas da marcha, entre as mulheres grávidas no segundo trimestre, mulheres não grávidas e mulheres com condições de sobrecarga artificiais; 3) analisar modelos biomecânicos com quatro set ups diferentes de análise; e, 4) analisar um modelo de contacto que determina a força vertical de reação do apoio. Os resultados demonstraram que as mulheres grávidas têm uma padrão de marcha similar ao normal. Observou-se que o ganho do peso no tronco aumenta o tempo das fases de apoio e de duplo apoio, quer nas mulheres grávidas quer nas mulheres com carga adicional. A resposta ao momento externo flexor da anca está relacionada com maior atividade dos extensores para suportar a carga anterior do tronco na direção da translação do centro de massa. Nas mulheres grávidas, o modelo universal-revolução-esférica afetou mais as variáveis cinemáticas quando comparado com o modelo de juntas com seis graus de liberdade. O modelo de contacto entre o pé e o solo, sobrestimou as forças verticais de reação. O aumento da massa do pé, devido ao inchaço consequente da gravidez, reduz a rigidez durante a fase de apoio. Os resultados do presente trabalho serão úteis para promover a investigação biomecânica do padrão de marcha durante a gravidez.
FCT - Fundação para Ciência e a Tecnologia

Dissertação de mestrado em Industrial Electronics and Computer Engineering
Several afflictions can affect a person’s ability to walk from muscular impairments, weakness or neurologic injury. In many of these cases, rehabilitation is essential for a full recovery. With the advances in the field of robotics and its bigger integration in rehabilitation, namely in the form of active orthosis and prosthesis, novel solutions to old challenges are made available. One of these challenges is the ability to use these assistive devices seamlessly without expert intervention in a subject’s daily life. Faced with this hindrance it becomes important to develop strategies that can recognize and predict human locomotion modes to allow a timely and correct response to a user’s needs from assistive devices. As such, this thesis proposes a pipeline of which the output is either a machine learning model that can recognize in real-time the user’s current locomotion mode or one that can predict a user’s walking intention. The locomotion mode recognition model can identify walking direction (forward, backward, anti-clockwise and clockwise) as well as locomotion activities (level walking, stair ascent, stair descent, ramp ascent and ramp descent) in real-time. Similarly, the intention prediction model also predicts both direction and locomotion activity intention in a timeframe that allows an assistive device to preemptively act in a seamless manner to provide the user a fluid walking ability and avoid a fall due to improper terrain traversing manner. An assessment of the required biomechanical features is done to identify the ones that best help predict or recognize the locomotion mode using feature selection methods (Principal Component Analysis, Analysis of variance-based selection, forward and backwards sequential selection). Several classification algorithms (Support Vector Machines, K-nearest neighbors, random forests and discriminant analysis) were explored and implemented to find the best performing one. These models were tested with data from healthy human subjects and a humanoid robot with a human-like gait controller. Results revealed that during the model building procedure using the Support Vector Machines algorithm with a feature selection method that combined the mRMR (minimum redundancy Maximum Relevancy) ranking technique and the forward feature selection procedure yielded the most robust and best-performing model. Direction prediction and recognition models presented an MCC (Matthews Correlation Coefficient) value of 0.98, on average, after validation showing promising results. However, and despite steady-state step type models (models classifying non-transitional steps) having an MCC value of 0.98, models involved in the classification of transitional steps, both for recognition and prediction, revealed poor results. MCC values as low as 0.61 were reported, showing that the used features were inadequate for the prediction of a subject’s gait intention. Future work will be to integrate other kinds of sensors and use different features that can rectify the classification flaws present in the obtained models in order to increase their accuracy.
Vários fatores podem afetar a capacidade de locomoção de uma pessoa desde lesões ou fraqueza musculares a lesões neurológicas. Em muito destes casos, reabilitação é essencial para uma completa recuperação. Com os avanços no campo da robótica e a sua maior integração em reabilitação, nomeadamente na forma de próteses e ortóteses ativas, novas soluções para velhos problemas tornam-se disponíveis. Um destes desafios é a habilidade de usar estes dispositivos assistivos de forma fluida e não-obstrutiva durante o dia-a-dia sem necessidade da intervenção de um especialista. Face a este problema torna-se importante desenvolver estratégias que possibilitem o reconhecimento e previsão de modos de locomoção humanos para permitir uma resposta correta e pontual de dispositivos assistivos face ás necessidades do utilizador. Como tal, esta tese propõe uma pipeline que tem como resultado um modelo de machine learning que consegue reconhecer em tempo-real o modo de locomoção enquanto acontece ou um modelo que consegue identificar a intenção de locomoção do utilizador. A pipeline delineada nesta tese permite obter um modelo que reconhece a direção de locomoção (frente, trás, anti-horário, horário) assim como a atividade locomotora (andar em terreno plano, subir escadas, descer escadas, subir rampas e descer rampas) em tempo real. A previsão de intenção também prevê tanto a direção como a atividade locomotora numa janela de tempo que permite ao dispositivo assistivo atuar sobre essa intenção. Um estudo das características biomecânicas necessárias é feito para identificar aquelas que melhor ajudam na previsão ou reconhecimento do modo de locomoção usando métodos de feature selection (Principal Component Analysis, ANOVA-based selection, forward e backwards sequential selection). Vários algoritmos de classificação (Support Vector Machines, K-nearest neighbors, Random Forests e Discriminant Analysis) foram explorados e implementados de forma a descobrir qual o melhor. Estes modelos foram testados com dados de sujeitos saudáveis e de um robô humanoide com um controlador de marcha humano. Resultados revelaram que para, a construção do modelo, o uso do algoritmo SVM e a seleção de features através da combinação dos métodos mRMR e forward feature selection resultavam no melhor e mais robusto modelo. Os modelos de classificação de direção, tato de reconhecimento como de previsão, obtiveram um valor de MCC de 0.98 em média depois da validação mostrando-se promissores. No entanto, e apesar dos modelos de reconhecimento e previsão de passos steady-state, terem obtido valores de MCC de 0.98, os modelos envolvidos na classificação de passos transicionais, tanto de reconhecimento como de previsão, obtiveram resultados fracos com valores de MCC tão baixos como 0.61, revelando que as features usadas são inadequadas para a previsão da intenção de marcha. Como trabalho futuro deverão ser integrados outros tios de sensores e usadas outras features que possam retificar as falhas de classificação presentes os modelos obtidos de forma a aumentar a sua perfromace.

Minimum toe clearance (MTC) above the walking surface is a critical representation of toe-trajectory control related to tripping risk. Reliable and precise MTC measurements are obtained in the laboratory using 3D motion capture technology. Real-world gait monitoring using body-mounted sensors presents considerable data processing challenges when estimating kinematic parameters, including MTC. This Thesis represents the first study employing machine-learning to estimate young and older adults’ toe-height at MTC using inertial data captured from a foot-mounted sensor. Age-group specific Generalized Regression Neural Network (GRNN) models estimated MTC with root-mean-square-error (RMSE) of 6.6 mm with 9 optimum inertial-signal features for the young and 7.1 mm with 5 features for the older during treadmill walking. These RMSE values are approximately one third of the previously reported (Mariani et al., 2012; McGrath et al., 2011) and GRNN modeling also performed well as reflected in no significant difference between 3D measured reference and model estimated MTC_Height. The GRNN model specific to older adults showed good generalizability when applied to data from slower and dual task walking.

PTPN22 encodes the lymphoid tyrosine phosphatase Lyp that can dephosphorylate Lck, ZAP-70 and Fyn to attenuate TCR signaling. A single-nucleotide polymorphism (C1858T) causes a substitution from arginine (R) to tryptophan (W) at 620 residue (R620W). Lyp-620W has been confirmed as a susceptible allele in multiple autoimmune diseases, including type 1 diabetes (T1D). Several independent studies proposed that the disease-associated allele is a gain-of-function variant. However, a recent report found that in human cells and a knockin mouse containing the R620W homolog that Ptpn22 protein degradation is accelerated, indicating Lyp-620W is a loss-of-function variant. Whether Lyp R620W is a gain- or loss-of-function variant remains controversial. To resolve this issue, we generated two lines (P2 and P4) of nonobese diabetic (NOD) mice in which Ptpn22 can be inducibly silenced by RNAi. We found long term silencing of Ptpn22 increased spleen cellularity and regulatory T (Treg) cell numbers, replicating the effect of gene deletion reported in the knockout (KO) B6 mice. Notably, Ptpn22 silencing also increased the reactivity and apoptotic behavior of B lymphocytes, which is consistent with the reduced reactivity and apoptosis of human B cells carrying the alleged gain-of-function PTPN22 allele. Furthermore, loss of Ptpn22 protected P2 KD mice from spontaneous and Cyclophosphamide (CY) induced diabetes. Our data support the notion that Lyp-620W is a gain-of-function variant. Moreover, Lyp may be a valuable target for the treatment of autoimmune diseases
PTPN22 kodiert die lymphoid tyrosine phosphatase Lyp, die Lck, ZAP-70 und Fyn dephosphorilieren kann, um T Zell Rezeptor Signale zu vermindern. Ein Polymorphismus (C1858T) verursacht einen Aminosäurenaustausch auf Position 620 von Arginin zu Tryptophan (R620W). Lyp-620W erhöht das Risiko einer Vielfalt von Autoimmunerkrankungen, darunter auch Diabetes Typ 1 (T1D). Mehrere Studien haben belegt, dass dieses Krankheitsallel die Funktion von Lyp verstärkt. Eine neuere Studie hat andererseits gezeigt, dass die R620W Variante schneller degradiert wird, was bedeuten würde, dass das C1858T Allel einen Funktionsverlust verursachen könnte. Ob Lyp R620W die Funktion dieser Phosphatase erhöht oder mindert bleibt demnach bis jetzt ungewiss. Um diese Frage zu klären haben wir zwei transgene Mauslinien (P2 und P4) im diabetischen Hintergrund der NOD Maus generiert, in denen Ptpn22 auf induzierbare Weise durch RNAi gehemmt werden kann. Unsere Ergebnisse zeigen, dass die langfristige Hemmung von Ptpn22 zu einer Zunahme der Milzzellularität und der Anzahl regulatorischer T Zellen führt, was dem Phänotyp des Ptpn22 knockout im B6 Hintergrund entspricht. Bemerkenswert ist, dass die Hemmung von Ptpn22 auch zu einer Zunahme der Reaktivität und des apoptotischen Verhaltens von B Lymphozyten führt, also dem entgegengesetzten Phänotypen, der in menschlichen B Zellen beobachtet wurde, die das Krankheitsallel exprimierten. Zusätzlich konnte die Ptpn22 Inhibierung NOD Mäuse vor spontanem und Cyclophosphamid-induziertem Diabetes schützen. Unsere Daten unterstützen also die Hypothese, dass Lyp-620W eine stärkere Aktivität vorweist. Dies würde auch bedeuten, dass Ptpn22 möglicherweise zu therapeutischen Zwecken inhibiert werden könnte, um Autoimmunerkrankungen zu bekämpfen

Mais do que uma organização bem estruturada e compacta, as empresas são compostas por biodiversidade humana, a qual constitui o código genético e cuja motivação e desempenho são tidos como essenciais para se distanciarem da concorrência. Distanciamento esse que pressupõe, igualmente, uma preocupação crescente centrada na melhoria contínua através da implementação de métodos e de processos inovadores como o Kaizen. Com efeito, o desenvolvimento de estratégias de melhoria contínua baseadas na filosofia Kaizen parece ser hoje uma corrente moderna comumente aceite na resposta aos novos desafios dos mercados. Partindo desta realidade, e com base no recurso a metodologias mistas, pretende-se com este trabalho, desenvolvido sobre a forma de estudo de caso, confirmar a hipótese da filosofia Kaizen poder contribuir positivamente para uma maior consciência e reforço de uma cultura de melhoria contínua em prol do cliente entre os colaboradores de um health club. Os resultados do estudo confirmam a hipótese, sugerindo que o Kaizen terá contribuído positivamente para um estado de consciência e de alerta maior para a questão da melhoria contínua e a sua importância ao nível dos processos de atendimento ao cliente e no reforço de uma cultura orientada para a melhoria do desempenho em prol do mesmo.
More than a well-structured and compact organization, companies are made up of human biodiversity, which constitutes the genetic code and whose motivation and performance are considered essential to distance themselves from the competition. This distancing also presupposes a growing concern centered on continuous improvement through the implementation of innovative methods and processes such as Kaizen. In fact, the development of continuous improvement strategies based on the Kaizen philosophy seems to be a modern current commonly accepted in the response to the new challenges of the markets. Starting from this reality, and based on the use of mixed methodologies, it is intended with this work, developed on the form of case study, to confirm the hypothesis of the Kaizen philosophy to contribute positively to a greater awareness and reinforcement of a culture of continuous improvement for the benefit of the client among the employees of a health club. The results of the study confirm the hypothesis, suggesting that Kaizen will have contributed positively to a state of awareness and greater alertness for the issue of continuous improvement and its importance in customer service processes and in strengthening a culture oriented towards the improvement of performance in favor of it.