Dissertations / Theses on the topic 'Electromyographie de surface (sEMG)'

Avec le vieillissement de la population, préserver la fonction musculaire est important pour éviter la perte de mobilité et d'autonomie. De nos jours, la prévention de la maladie musculaire, la sarcopénie, est une préoccupation majeure et des facteurs de risque importants tels que l'âge avancé ainsi que des facteurs modifiables, notamment une faible activité physique et une alimentation déséquilibrée ont été identifiés. Compte tenu de la croissance des populations plus âgées et de la diminution de l'activité physique, qui touche également les jeunes citoyens, la sensibilisation à la qualité musculaire peut être cruciale pour promouvoir un vieillissement en bonne santé dans nos sociétés. Les besoins en évaluations fonctionnelles musculaires ont été exprimés par les chercheurs et les cliniciens. Le groupe de travail européen sur la sarcopénie chez les personnes âgées (EWGSOP) recommande de définir la sarcopénie comme la présence à la fois d'une faible masse musculaire et d'une faible fonction musculaire (force et performance physique). Pour cela, nous avons développé une méthode d’évaluation du vieillissement musculaire, en utilisant une technologie ambulatoire et non invasive, appelée technologie d'électromyographie de surface haute densité (HD-sEMG), à travers un projet de recherche clinique sur cinq catégories d'âge (25 à 74 ans), actifs et sédentaires. Nous avons réalisé une étude comparative avec une analyse complète et multimodale du rectus femoris (RF), muscle impliqué dans les mouvements de la vie quotidienne, pour dévoiler le potentiel prometteur de la technique HD sEMG, par rapport aux techniques cliniques classiques, l’objectif étant de détecter les changements précoces de la qualité de la fonction musculaire impactée par le vieillissement et le niveau d'activité physique. La partie clinique de ce projet de thèse a été financée par une subvention européenne, EIT Health. En analysant principalement la dynamique de contraction musculaire et l'intensité du rectus femoris, nos résultats ont montré que la technique HD-sEMG, était capable de discriminer entre les cinq catégories d'âge de sujets sains physiquement actifs. Plus intéressant encore, les scores HD-sEMG proposés discriminaient entre les participants actifs et sédentaires, de la même catégorie d'âge (45-54 ans), contrairement aux paramètres cliniques et aux autres techniques couramment utilisées (absortiométrie biphotonique par rayons X, DXA et échographie). De plus, ces scores pour les participants sédentaires de cette catégorie d'âge étaient significativement plus proches de ceux des participants actifs des catégories d'âge supérieures (55-64 ans et 65-74 ans). Cela suggère fortement qu'un mode de vie sédentaire semble accélérer le processus de vieillissement musculaire au niveau anatomique et fonctionnel, et ce processus accéléré subtil peut être détecté par la technique HD-sEMG. Ces résultats préliminaires prometteurs pourraient contribuer au développement d’un outil intéressant aux cliniciens pour améliorer à la fois la précision et la sensibilité de l'évaluation musculaire utile pour les programmes de prévention et de réadaptation afin d'éviter ou de retarder la sarcopénie, problème de santé publique actuel alerté par l'Organisation Mondiale de la Santé (OMS) et promouvoir un vieillissement en bonne santé
With the aging of the population, preserving muscle function is important to prevent loss of mobility and autonomy. Nowadays, the prevention of the muscle disease, sarcopenia, is a major concern and important risk factors such as older age as well as modifiable factors including low physical activity and unhealthy diet have been identified. Considering the growth of older populations and the decreased physical activity, which also includes young citizens, muscle quality awareness can be crucial in promoting a healthy aging process in our societies. Muscle functional assessments needs were expressed by researchers and clinicians, The European Working Group on Sarcopenia in Older People (EWGSOP) recommends defining sarcopenia as the presence of both low muscle mass and low muscle function (strength, and physical performance). For this purpose, we have developed a method for muscle aging evaluation, using an ambulatory and non-invasive technology, called high-density surface electromyography (HDsEMG), through a clinical research project on five age categories (25 to 74 yrs.). We performed a comparative study with a complete and multimodal analysis of the rectus femoris, muscle involved in daily life motions, in order to reveal the promising potential of the HD-sEMG technique, compared to conventional clinical techniques, to detect early changes in the quality of muscle function impacted by aging and physical activity level. The clinical part of this thesis project was funded by a European grant, EITH Health. By analyzing both muscle contraction dynamics and intensity of the rectus femoris, our results showed that the HD-sEMG technique, was able to discriminate between the five age categories of healthy physically active subjects. More interestingly, the proposed HD-sEMG scores discriminated between active and sedentary participants, from the same age category(45-54 yrs.), in contrary to clinical parameters and others usual techniques (dual-energy x-ray absorptiometry, DXA and ultrasonography). In addition, these scores for sedentary participants from this age category were significantly closer to those of active participants from higher age categories (55-64 yrs. and 65-74 yrs.). This strongly suggests that sedentary lifestyle seems to accelerate the muscle aging process at both anatomical and functional level, and this subtle accelerated process can be detected by the HD-sEMG technique. These promising preliminary results can contribute to the development of an interesting tool for clinicians to improve both accuracy and sensitivity of functional muscle evaluation useful for prevention and rehabilitation to avoid the effects of unhealthy lifestyle that can potentially lead to sarcopenia. This can support also the actual public health concern alerted by Word Health Organization (WHO) regarding aging and sarcopenia, to promote healthy aging

Le vieillissement musculaire, en tant qu'entité pathologique, est connu sous le nom de sarcopénie. Il est défini comme une réduction de la force musculaire accompagnée d'une perte de masse musculaire et d'un déclin des fonctions physiques. Les méthodologies actuellement utilisées en pratique clinique pour évaluer cette maladie liée au vieillissement sont plutôt limitées pour saisir les caractéristiques de ce déclin à l'échelle macroscopique: mesure de la force et de la masse musculaire. Cependant, diagnostiquer la sarcopénie en mesurant uniquement ces deux paramètres n'est pas assez précis et ne permet pas de détecter une perte précoce de la fonction musculaire. Il est plus fiable d'exploiter des changements musculaires à l’échelle microscopique: tels que la perte des unités motrices (l'unité motrice (UM) est constituée d'un motoneurone et de toutes les fibres musculaires squelettiques innervées par ses terminaisons axonales), l'atrophie des fibres, le désordre de la commande neuronale, et l’infiltration intramusculaire des cellules adipeuses. Ainsi, des études récentes, basées sur la technique d'électromyographie de surface (sEMG), ont démontré le grand potentiel de cette technique en tant que biomarqueur pour détecter les premiers signes de muscles sarcopéniques. En effet, le signal sEMG est la réponse électrique de l'activation musculaire gérée par le système nerveux central (SNC). Il est mesuré de manière non invasive à la surface de la peau à l'aide d'électrodes de surface et peut être efficacement corrélé à la réponse mécanique de l'activation musculaire. De plus, les modèles mathématiques du signal sEMG peuvent former une alliance utile avec les mesures expérimentales et le traitement du sEMG pour identifier et/ou quantifier les bio-indicateurs (c'est-à-dire les paramètres anatomiques et neuronaux des muscles) d'un vieillissement musculaire sain, précoce, accéléré ou sarcopénique. Dans cette thèse, nous avons utilisé un modèle électrique optimisé décrivant l'activité électrique du muscle à la surface de la peau à l'aide de la technique d'électromyographie à haute densité (HD-sEMG). Ce modèle est développé précédemment dans notre laboratoire de recherches. Le temps de calcul réduit de ce modèle est l'élément clé majeur pour effectuer l'identification des indicateurs de vieillissement à l'aide de méthodes inverses et de la technique HD-sEMG. Cependant, cette identification nécessite des méthodes préalables telles que l'analyse de sensibilité et d'identifiabilité. De plus, lors de l'utilisation de ce modèle, nous avons observé d'importantes limitations telles qu'un manque de réalisme physiologique (par exemple, les territoires des UM et le nombre de fibres par muscle), de personnalisation (par exemple, le même schéma de recrutement neuronal pour les sujets jeunes et âgés) et de simplicité (par exemple, l'ajustement de 50 paramètres de modèle en fonction de l'âge et du sexe). Ces limitations restreignent l'utilisation de ce modèle dans le diagnostic du vieillissement musculaire. Par conséquent, l'objectif de cette thèse est de remédier à ces limitations et de fournir un modèle plus réaliste et simple à utiliser pour évaluer le vieillissement musculaire. Dans ce travail, nous proposons d'abord une méthode d’analyse de sensibilité de Morris améliorée (IMSA) appliquée au modèle développé. Cette analyse a été réalisée sur des sujets simulés jeunes et âgés (à faible et à fort niveau de contractions). Grâce à cette IMSA, nous avons réussi à mettre en évidence et avec précision les paramètres/facteurs neuromusculaires influents pour chaque catégorie d'âge, à chaque niveau de force et pour chaque descripteur statistique calculée à partir des signaux HD-sEMG. De plus, grâce à l'IMSA, nous avons mis en évidence les limitations du modèle mentionnées précédemment. Pour remédier à ces limitations, nous avons modifié le schéma du modèle pour le rendre plus facile à manipuler, avec moins de risques d'erreurs et d'incohérences
The muscle aging, as a disease entity, is known as Sarcopenia. It is defined as a reduction of muscle strength/force accompanied by a loss of muscle mass and a decline in physical functions. The current methodologies used in clinical practice to assess this aging disease, are rather limited to capture the features of this decline at the macroscopic scale. Factors such as the loss of Motor Units (motor unit (MU) is made up of a motoneuron and all the skeletal muscle fibers innervated by the neuron's axon terminals), the atrophy of fibers and the disorder of the neural recruitment pattern are shown to have a clear influence on muscular function. However, diagnosing sarcopenia by only measuring the muscle strength and/or muscle mass is not enough accurate and cannot alert an early loss of muscular function. The inner scales (MU and fiber scale age-related changes) reflecting that loss of muscle mass and strength during aging are more interesting to exploit. Thus, recent studies, based on the surface electromyography (sEMG) technique, have demonstrated the great potential of this technique to be used as a biomarker to detect early signs of sarcopenic muscles. In fact, the sEMG signal is the electrical response of the muscle activation managed by the Central Nervous System (CNS). It is measured with a noninvasive manner at the skin surface using surface electrodes and can be correlated efficiently to the mechanical response of muscle activation. Moreover, mathematical models of sEMG signal can form a useful alliance with sEMG experimental measures and processing to identify and/or quantify bio-indicators (i.e., anatomical, and neural muscle parameters) of a healthy, early, accelerated or sarcopenic muscle aging. In this thesis work, we have used a fast and optimized electrical model describing the electrical activity of the muscle at the skin surface using High Density sEMG technique (HD-sEMG), developed in our laboratory team. The reduced computational time of this model is the major key feature to perform the identification of aging indicators using inverse methods and HD-sEMG technique. However, this identification needs pre-aided-methods such as the sensitivity and the identifiability analysis. Moreover, when dealing with this model, we have observed important limitations such as lack of physiological realism (e.g., MUS territories and the number of fibers per muscle), personalization (e.g., same recruitment pattern for young and elder subject), and simplicity (e.g., adjustment of 50 model parameters according to age and gender). These limitations restrain the use of this model in muscle aging diagnosis. Therefore, we aimed in this thesis to address the limitations of this model and deliver more realistic and user-friendly model to evaluate muscle aging. Therefore, in this work, we first propose an Improved Morris Sensitivity Analysis (IMSA) applied on the developed model. This analysis was performed on young and elder simulated subjects (at low and high force level). Using this IMSA, we success to spotlight with accuracy the influential neuromuscular parameters/factors for each age category, at each force level, and for each statistic feature computed over the HD-sEMG signals. Furthermore, using IMSA, we have outlined the model inaccuracies and limitations mentioned above. To address these limitations, we have modified the model schema implementation to be easier to manipulate (user-friendly model), with less error and inconsistency risks. Only the age and the gender of subject became needed as model entries to initiate a simulation of HD-sEMG signals. All other parameters necessary in simulations are then estimated through "statistical" models. The statistical models employ regression analysis to estimate the relation Parameter versus Age. A bibliographic research reporting these morphological and structural changes according to age, gender, and Biceps Brachii muscle was done

Hand gesture recognition based on surface electromyographic (sEMG) signals is a promising approach for the development of Human-Machine Interfaces (HMIs) with a natural control, such as intuitive robot interfaces or poly-articulated prostheses. However, real-world applications are limited by reliability problems due to motion artifacts, postural and temporal variability, and sensor re-positioning. This master thesis is the first application of deep learning on the Unibo-INAIL dataset, the first public sEMG dataset exploring the variability between subjects, sessions and arm postures, by collecting data over 8 sessions of each of 7 able-bodied subjects executing 6 hand gestures in 4 arm postures. In the most recent studies, the variability is addressed with training strategies based on training set composition, which improve inter-posture and inter-day generalization of classical (i.e. non-deep) machine learning classifiers, among which the RBF-kernel SVM yields the highest accuracy. The deep architecture realized in this work is a 1d-CNN implemented in Pytorch, inspired by a 2d-CNN reported to perform well on other public benchmark databases. On this 1d-CNN, various training strategies based on training set composition were implemented and tested. Multi-session training proves to yield higher inter-session validation accuracies than single-session training. Two-posture training proves to be the best postural training (proving the benefit of training on more than one posture), and yields 81.2% inter-posture test accuracy. Five-day training proves to be the best multi-day training, and yields 75.9% inter-day test accuracy. All results are close to the baseline. Moreover, the results of multi-day trainings highlight the phenomenon of user adaptation, indicating that training should also prioritize recent data. Though not better than the baseline, the achieved classification accuracies rightfully place the 1d-CNN among the candidates for further research.

The force produced by a specific muscle cannot be measured and what is measured is the total force provided by all the active muscles acting on a joint, including agonists and antagonists. The first part of this work (chapter 3) addresses the issue of load sharing by proposing two possible approaches and testing them. The second part (chapter 4 and 5) addresses two applications of surface EMG focusing on the study of a) muscle relaxation associated to Yoga sessions and b) the activation of muscle of the back and shoulder of musicians playing string instruments (violin, viola and cello). In both parts the element of innovation is the use of two dimensional electrode arrays and of techniques based on EMG Imaging. The objectives of this work are presented and explained in chapter 1 while the basic concepts of surface EMG are summarized in chapter 2. Different EMG-based muscle force models found in the literature are explained and discussed. Two renowned amplitude indicators in surface EMG (sEMG) studies are the average rectified value (ARV) and the root mean square (RMS). These two amplitude indicators are computed over a defined time window of the recorded signals to represent the muscle activity. The advantages and disadvantages of RMS and ARV are compared and discussed for a simple sinusoid as well as for more complex signals (simulated motor unit action potential detected by high density electrode grid). The results show that RMS is more robust to the sampling frequency than ARV. In this thesis, starting from the simulation of a single fiber and of a group of fibers (motor unit), it is shown that inter electrode distance (IED) greater than10 mm causes aliasing. Aliasing is a source of error in sEMG map interpretation or decisions that are made by automatic algorithms such as those providing image segmentation for the identifications of regions of interest. Chapter 2 discusses three segmentation algorithms (K-means, h-dome, watershed) and compares them in order to find the most suitable method. Results reveal that among the three mentioned algorithms, watershed provides most accurate segmentation for the simulated ARV maps. Chapter 3 presents a mathematical model that is associated to the monotonic Force-EMG relation. A possible non-linear relationship between the EMG and force or torque is presented. A system of "M" equations is obtained by performing "M" measurements at "M" different force levels in isometric conditions. The solutions of such system of equations are the values for each muscles. Two different approaches were investigated for finding the solutions of the system, which are: a) Analytical-Graphical Approach (AGA) and b) Numerical Approach (NA) consisting of error minimization (between the total estimated and measured force) applying optimization algorithms. The AGA was used to find the model parameters of each muscle contributing to the force production on a joint by finding the intersection of those surfaces that can be obtained from sequential substitutions of the model parameters in the equations corresponding to each contraction level. In simulation studies, the AGA graphically shows that there is more than one solution to the load sharing problem even for the simplest theoretical case (i.e. a joint spanned by only two muscles). The second approach, based on minimization of the mean square error between the measured and the total estimated force or torque (with "N" muscles involved) provides an estimate of the model parameters that in turn provides the force contributions of the individual muscles. The optimization algorithms can find the solutions of our system made of non-linear equations (see chapter 3). Starting from different point (initial conditions), different solutions can be found, as predicted by the AGA approach for the two-muscle case. The main conclusion of this study is that the load sharing strategy is not unique. Chapter 4 discusses the application of surface electromyography to a single case study of Yoga relaxation to show the feasibility of measurements. The effect of yoga relaxation on muscle activity (sEMG amplitude), as well as on mean and median frequencies and muscle fiber's conduction Velocity, is discussed in this chapter. No changes in the sEMG activity pattern distribution were found for the same task performed before and after applying yoga relaxation technique. However, myoelectric manifestations of fatigue were smaller after relaxation and returned to the normal pattern after the recovery phase from relaxation. Further studies are justified. Chapter 5 describes results and discusses the spatial distribution of muscle activity over the Trapezius and Erector Spinae muscles of musicians playing string instruments. In chapter 5, the effect of backrest support in sitting position during playing cello, viola, and violin on the muscle activity index of upper and lower Trapezius muscle of the bowing arm, upper Trapezius muscle of non-bowing arm, left and right Erector Spinae muscles is investigated. Two professional players (one cello and one viola) and five student players (one cello, three violin and one viola) participated in this study. The muscle activity index (MAI) was defined as the spatial average of RMS values of the muscle active region detected by watershed segmentation for Trapezius muscles (left and right), and thresholding technique (70% of the maximum value) for left and right Erector Spinae muscles. It was found that the MAI is string (note) dependent. Statistical difference (p < 0:05) between the MAIs of left Erector Spinae muscle during playing with and without backrest support was observed in four (out of five) student players. No statistical differences were observed on the muscle activity of Trapezius (bowing and no-bowing arms) during playing with and without backrest support in different types of bowing for all musicians. In conclusion, this work addresses a) the issue of spatial sampling and segmentation of sEMG using 2D electrode arrays, b) two possible approaches to the load-sharing issue, c) a single-case study of Yoga relaxation and d) the distribution of muscle activity above the Trapezius and Erector Spinae muscles of musicians playing string instruments. Previously unavailable knowledge has been achieved in all these four studies.

ABSTRACT Objectives: Swallowing is a complex physiologic function developing mostly in the first years of life. After 6 years old, if a complete maturation is not achieved, swallowing persists as “atypical swallowing” (AS). The therapy provided to re-educate this dysfunction is based on the myofunctional treatment (MFT). The aim of this study was to detect functional (electromyographical) and clinical (orofacial muscular evaluation with score (OMEs) protocol) effects of MFT in a group of patients with AS so to highlight any differences in the muscular activation pattern and muscular orofacial behavior. Materials and Methods: 20 adolescents and young adults (4 males and 16 females, mean age 17.85 years, SD 4.80) with AS were selected for this study. Standardized surface electromyographic (ssEMG) analysis was performed by the same operator to detect the activity of masseter (MM), temporalis (TA) and submental (SM) muscles before (T1) and after (T2) the logopedic treatment. The MFT was performed by the same speech therapist according to the Garliner method for a period of 10 weeks. The speech therapist completed the OMEs protocol at T1 and T2. A Student-t test for paired data was carried out to detect differences between T1 and T2 for both ssEMG and OMEs data. Then, a 1-way ANOVA variance test was performed to detect any differences between the different couples of muscles at T1 and T2. In addition, ssEMG data at T1 and T2 were compared with ssEMG obtained in a control (C) group of 18 adolescents and young adult patients (8 males and 10 females, mean age 17.28 years, SD 2.56) with bimolar class 1 and without AS. Results: From the starting group of 20 patients, 15 patients completed the MFT (4 males and 11 females, mean age 17.72 years, SD 5.21). At T2, AS patients showed a significantly shorter duration of activation for each couple of muscles and for the whole duration act of swallowing (p<.0001) as well as higher intensity of the SM activity (p<.01) than at T1. Within the AS group, at T1 the masticatory muscles (MM and TA) showed lower duration of activation (p<.05) and lower intensity of the spike (p<.0001) than SM. At T2, masticatory muscles also showed lower values for the activation index (IMPACT) (p<.0001) and for the spike position (p<.01) than SM. At T2. The OMEs protocol showed a significant increase for the total evaluation (p<.01) and specifically for appearance and posture (p<.01) and functions (p<.0001). If compared to C group, the AS group at T1 showed significantly longer duration of activation for each couple of muscles and for the whole duration act of swallowing (p<.0001) as well as lower intensity of the SM activity (p<.05) than controls. At T2 all the ssEMG data detected in AS patients showed a general improvement and moved toward the control values. The differences between AS and C groups about the duration of activation of each couple of muscles and the whole duration act of swallowing were lower at T2 than at T1 even if still significantly different from C ones (p<.0001). Conclusion: MFT confirms itself as an effective method in the treatment of AS dysfunction permitting a shortening of the muscular activation pattern, an increase in SM activity and a general improvement in the orofacial muscular behavior making them closer to the data obtained in controls. ssEMG and OMEs protocol represent valid and useful methods in the analysis of the swallowing function and in establishing the effects of the MFT.

Industrial robots are delivering more and more manipulation services in manufacturing. However, when the task is complex, it is difficult to programme a robot to fulfil all the requirements because even a relatively simple task such as a peg-in-hole insertion contains many uncertainties, e.g. clearance, initial grasping position and insertion path. Humans, on the other hand, can deal with these variations using their vision and haptic feedback. Although humans can adapt to uncertainties easily, most of the time, the skilled based performances that relate to their tacit knowledge cannot be easily articulated. Even though the automation solution may not fully imitate human motion since some of them are not necessary, it would be useful if the skill based performance from a human could be firstly interpreted and modelled, which will then allow it to be transferred to the robot. This thesis aims to reduce robot programming efforts significantly by developing a methodology to capture, model and transfer the manual manufacturing skills from a human demonstrator to the robot. Recently, Learning from Demonstration (LfD) is gaining interest as a framework to transfer skills from human teacher to robot using probability encoding approaches to model observations and state transition uncertainties. In close or actual contact manipulation tasks, it is difficult to reliabley record the state-action examples without interfering with the human senses and activities. Therefore, wearable sensors are investigated as a promising device to record the state-action examples without restricting the human experts during the skilled execution of their tasks. Firstly to track human motions accurately and reliably in a defined 3-dimensional workspace, a hybrid system of Vicon and IMUs is proposed to compensate for the known limitations of the individual system. The data fusion method was able to overcome occlusion and frame flipping problems in the two camera Vicon setup and the drifting problem associated with the IMUs. The results indicated that occlusion and frame flipping problems associated with Vicon can be mitigated by using the IMU measurements. Furthermore, the proposed method improves the Mean Square Error (MSE) tracking accuracy range from 0.8˚ to 6.4˚ compared with the IMU only method. Secondly, to record haptic feedback from a teacher without physically obstructing their interactions with the workpiece, wearable surface electromyography (sEMG) armbands were used as an indirect method to indicate contact feedback during manual manipulations. A muscle-force model using a Time Delayed Neural Network (TDNN) was built to map the sEMG signals to the known contact force. The results indicated that the model was capable of estimating the force from the sEMG armbands in the applications of interest, namely in peg-in-hole and beater winding tasks, with MSE of 2.75N and 0.18N respectively. Finally, given the force estimation and the motion trajectories, a Hidden Markov Model (HMM) based approach was utilised as a state recognition method to encode and generalise the spatial and temporal information of the skilled executions. This method would allow a more representative control policy to be derived. A modified Gaussian Mixture Regression (GMR) method was then applied to enable motions reproduction by using the learned state-action policy. To simplify the validation procedure, instead of using the robot, additional demonstrations from the teacher were used to verify the reproduction performance of the policy, by assuming human teacher and robot learner are physical identical systems. The results confirmed the generalisation capability of the HMM model across a number of demonstrations from different subjects; and the reproduced motions from GMR were acceptable in these additional tests. The proposed methodology provides a framework for producing a state-action model from skilled demonstrations that can be translated into robot kinematics and joint states for the robot to execute. The implication to industry is reduced efforts and time in programming the robots for applications where human skilled performances are required to cope robustly with various uncertainties during tasks execution.

L'identification rapide ou en temps réel de l'activation spatio-temporelle des unités motrices (UM) qui représentent les unités fonctionnelles du système neuromusculaire est fondamentale dans les applications de contrôle des prothèses et en réhabilitation fonctionnelle. Cependant, cette procédure demande un temps de calcul énorme. Par conséquent, le travail de cette thèse a été consacré à fournir un algorithme permettant l'identification en temps réel des stratégies d'activation spatiale et temporelle des UMs en appliquant des méthodes inverses sur les signaux HD-sEMG (électromyogramme de surface à haute densité) à partir d'une grille placée sur le Biceps Brachial (BB). À cette fin, nous proposons une approche innovante, qui implique l'utilisation de la méthode inverse classique de minimisation de norme et une interpolation de courbe en 3D, à savoir l'approche est nommée CFB-MNE. Cette méthode, fondée sur l'identification inverse (estimation de la norme minimale) couplée à un dictionnaire des potentiels d'action des unités motrices simulées (MUAP) d'un modèle récent et testée sur des simulations, a permis la localisation en temps réel des unités motrices individuelles simulées. Une analyse de robustesse (modifications anatomiques, physiologiques et instrumentales) a ensuite été effectuée pour vérifier l'efficacité de l'algorithme proposé. Enfin, l'algorithme proposé a été testé sur des UMs avec des schémas de recrutement réalistes donnant des résultats prometteurs et encourageants en identification spatiale et temporelle sur trois scenarios. Pour conclure, en perspectives, les résultats prometteurs obtenus suggèrent l'utilisation de l'apprentissage automatique et de l'intelligence artificielle (IA) pour améliorer encore les performances de l'algorithme proposé
Fast or real-time identification of the spatiotemporal activation of Motor Units (MUs), functional units of the neuromuscular system, is fundamental in applications as prosthetic control and rehabilitation guidance but often dictates expensive computational times. Therefore, the thesis work was devoted to providing an algorithm that enables the real-time identification of MU spatial and temporal activation strategies by applying inverse methods on HD-sEMG (high-density surface electromyogram) signals from a grid placed over the Biceps Brachii (BB). For this purpose, we propose an innovative approach, that involves the use of the classical minimum norm inverse method and a 3D fitting curve interpolation, namely CFB-MNE approach. This method, based on inverse identification (minimum norm estimation) coupled to simulated motor unit action potential (MUAP) dictionary from a recent model and tested on simulations, allowed the real time localization of simulated individual motor units. A robustness analysis (anatomical, physiological, and instrumental modifications) was then performed to verify the efficiency of the proposed algorithm. Finally, the proposed algorithm was tested on MUs with realistic recruitment patterns giving promising results in both spatial and temporal identification. To conclude, a door to future perspectives was opened, according to the obtained promising results, suggesting the use of machine learning and artificial intelligence (AI) to further boost the performance of the proposed algorithm

This thesis has evaluated the use of Independent Component Analysis (ICA) on Surface Electromyography (sEMG), focusing on the biosignal applications. This research has identified and addressed the following four issues related to the use of ICA for biosignals: • The iterative nature of ICA • The order and magnitude ambiguity problems of ICA • Estimation of number of sources based on dependency and independency nature of the signals • Source separation for non-quadratic ICA (undercomplete and overcomplete) This research first establishes the applicability of ICA for sEMG and also identifies the shortcomings related to order and magnitude ambiguity. It has then developed, a mitigation strategy for these issues by using a single unmixing matrix and neural network weight matrix corresponding to the specific user. The research reports experimental verification of the technique and also the investigation of the impact of inter-subject and inter-experimental variations. The results demonstrate that while using sEMG without separation gives only 60% accuracy, and sEMG separated using traditional ICA gives an accuracy of 65%, this approach gives an accuracy of 99% for the same experimental data. Besides the marked improvement in accuracy, the other advantages of such a system are that it is suitable for real time operations and is easy to train by a lay user. The second part of this thesis reports research conducted to evaluate the use of ICA for the separation of bioelectric signals when the number of active sources may not be known. The work proposes the use of value of the determinant of the Global matrix generated using sparse sub band ICA for identifying the number of active sources. The results indicate that the technique is successful in identifying the number of active muscles for complex hand gestures. The results support the applications such as human computer interface. This thesis has also developed a method of determining the number of independent sources in a given mixture and has also demonstrated that using this information, it is possible to separate the signals in an undercomplete situation and reduce the redundancy in the data using standard ICA methods. The experimental verification has demonstrated that the quality of separation using this method is better than other techniques such as Principal Component Analysis (PCA) and selective PCA. This has number of applications such as audio separation and sensor networks.

L’objectif de cette thèse a été d’analyser l’effet de l’exercice physique réalisé sur plateforme vibrante (whole-body vibration, WBV) sur l’activité musculaire des membres inférieurs, de développer des outils d’analyse méthodologiques et de proposer des recommandations pratiques d’utilisation. Deux études méthodologiques ont été menées pour identifier la méthode optimale permettant de traiter les signaux d'électromyographie de surface (sEMG) recueillis pendant la vibration et d'analyser l'influence de la méthode de normalisation de l'activité sEMG. Une troisième étude visait à mieux comprendre si les pics sEMG observés dans le spectre de puissance du signal contiennent des artéfacts de mouvement et/ou de l'activité musculaire réflexe. Les trois études suivantes avaient pour but de quantifier l’effet de la WBV sur l’activité musculaire en fonction de différents paramètres tels que, la fréquence de vibration, l'amplitude de la plateforme, une charge supplémentaire, le type de plateforme, l'angle articulaire du genou, et la condition physique du sujet. En outre, l'objectif a été de déterminer l'accélération verticale minimale permettant de stimuler au mieux l'activité musculaire des membres inférieurs. En résumé, les recherches menées au cours de cette thèse fournissent des solutions pour de futures études sur : i) comment supprimer les pics dans le spectre du signal sEMG et, ii) comment normaliser l'activité musculaire pendant un exercice WBV. Enfin, les résultats de cette thèse apportent à la littérature scientifique de nouvelles recommandations pratiques liées à l’utilisation des plateformes vibrantes à des fins d’exercice physique
The aim of this thesis was to analyze the effect of whole-body vibration (WBV) exercise on lower limb muscle activity and to give methodological implications and practical applications. Two methodological studies were conducted that served to evaluate the optimal method to process the surface electromyography (sEMG) signals during WBV exercise and to analyze the influence of the normalization method on the sEMG activity. A third study aimed to gain insight whether the isolated spikes in the sEMG spectrum contain motion artifacts and/or reflex activity. The subsequent three investigations aimed to explore how the muscle activity is affected by WBV exercise, with a particular focus on the vibration frequency, platform amplitude, additional loading, platform type, knee flexion angle, and the fitness status of the WBV user. The final goal was to evaluate the minimal required vertical acceleration to stimulate the muscle activity of the lower limbs. In summary, the research conducted for this thesis provides implication for future investigations on how to delete the excessive spikes in the sEMG spectrum and how to normalize the sEMG during WBV. The outcomes of this thesis add to the current literature in providing practical applications for exercising on a WBV platform

Hamstring strain injuries continue to be expensive and time consuming for running based sports. Of the hamstring group, the biceps femoris long head is the most often injured. To date, hamstring injury rehabilitation and prevention methods have focussed predominantly on knee flexion exercises, despite the fact that hip extension exercises better activate this muscle. The aim of this thesis was to compare hamstring activation during variants of the hip extension based Romanian deadlift, to explore the patterns of hamstring and other hip extensor muscle activation. Data from this thesis can be used as a basis for future training studies.

L'objectif de cette these est de decrire la locomotion du cheval au trot grace a l'utilisation conjointe de l'electromyographie de surface, de l'accelerometrie et de la cinematique. L'introduction propose une revue des differentes methodes d'etude de la locomotion employees chez le cheval. La deuxieme partie demontre que l'electromyographie de surface peut etre utilisee sans difficultes majeures chez cet animal. Son association avec la cinematique articulaire permet de determiner le plan d'activation et la fonction de la plupart des plus volumineux muscles superficiels du tronc et des membres. Les troisieme et quatrieme parties appliquent la methode mise en place a l'etude du trot du cheval au cours de quelques exercices utilises pour son entrainement : allongements d'allure, travail sur plan incline et abaissement de l'encolure. Elles demontrent que des changements de vitesse de l'ordre de 0,5 m/s ou des variations de la pente de 3 o induisent des modifications significatives des periodes et de l'intensite de l'activite musculaire et des decours articulaires au cours du cycle de la foulee. L'abaissement de l'encolure modifie les mouvements du tronc mais influe peu sur les membres. La derniere partie analyse les apports de ce travail. Elle en presente les limites et les perspectives au regard de l'evolution actuelle de l'etude de l'activite musculaire chez le cheval.

L'estimation de la force générée par un muscle est importante dans les études biomécaniques et pour les applications cliniques. Puisque cette force ne peut pas être mesurée directement, le signal électromyographique de surface (SEMG), reflétant le niveau d'activation musculaire, est utilisé pour quantifier la force développée. Cependant, tous les facteurs, contrôlant une contraction isométrique, n'influencent pas la force et le SEMG simultanément. Le but de ce travail de thèse est donc de développer un modèle de simulation conjointe du SEMG et de la force, afin d'étudier la relation EMG-force. Dans ce but, nous avons d'abord développé une nouvelle méthode de simulation de la force musculaire à partir d'un modèle d'EMG existant. Le modèle complet a été testé pour le choix de la stratégie de recrutement et l'influence de la durée de la consigne. Puis, nous avons utilisé une méthode de Monte Carlo pour étudier la sensibilité du modèle aux différents paramètres physiologiques d'entrée. Deux critères existants (relations EMG-force et force-variabilité de force) ainsi qu'un nouveau critère (erreur entre la consigne de force et la force générée), ont été utilisés pour optimiser les paramètres avec une consigne de force constante. Ce nouveau critère a ensuite été utilisé avec une consigne de force variable (sinusoïdale ou triangulaire), afin d'obtenir les plages optimales des paramètres. Enfin, pour évaluer notre modèle, nous avons réalisés des expérimentations et une simulation pour le biceps. Les résultats montrent que notre modèle EMG-force est capable de simuler qualitativement les comportements réels du biceps pour les contractions isotoniques et anisotoniques.

The relationship between sEMG signals and muscle force, and associated joint torque, is an object of study for clinical applications such as rehabilitation robotics and commercial applications as wearable motion control devices. The information type and quality obtained by sEMG can impact the classification and prediction accuracy of ankle joint torque. In this thesis project, HD-sEMG based data was collected together with ankle joint torque measurements from 5 subjects during MVIC of plantarflexors and dorsiflexors. Machine learning approaches ideally suited for nonlinear regression tasks, such as MLP and LSTM, have been implemented and evaluated to best predict joint torque profiles given extracted features from sEMG data. An evaluation of machine learning performances using HD-sEMG data over bipolar sEMG data has been conducted in intra-session, inter-subjective and intra-subjective study cases.

Les systèmes neuromusculaires et musculo-squelettique sont considérés comme un système de systèmes complexe. En effet, le mouvement du corps humain est contrôlé par le système nerveux central par l'activation des cellules musculaires squelettiques. L'activation du muscle produit deux phénomènes différents : mécanique et électrique. Ces deux activités possèdent des propriétés différentes, mais l'activité mécanique ne peut avoir lieu sans l'activité électrique et réciproquement. L'activité mécanique de la contraction du muscle squelettique est responsable du mouvement. Le mouvement étant primordial pour la vie humaine, il est crucial de comprendre son fonctionnement et sa génération qui pourront aider à détecter des déficiences dans les systèmes neuromusculaire et musculo-squelettique. Ce mouvement est décrit par les forces musculaires et les moments agissant sur une articulation particulière. En conséquence, les systèmes neuromusculaires et musculo-squelettique peuvent être évalués avec le diagnostic et le management des maladies neurologiques et orthopédiques à travers l'estimation de la force. Néanmoins, la force produite par un seul muscle ne peut être mesurée que par une technique très invasive. C'est pour cela, que l'estimation de cette force reste l'un des grands challenges de la biomécanique. De plus, comme dit précédemment, l'activation musculaire possède aussi une réponse électrique qui est corrélée à la réponse mécanique. Cette résultante électrique est appelée l'électromyogramme (EMG) et peut être mesurée d'une façon non invasive à l'aide d'électrodes de surface. L'EMG est la somme des trains de potentiel d'action d'unité motrice qui sont responsable de la contraction musculaire et de la génération du mouvement. Ce signal électrique peut être mesuré par des électrodes à la surface de la peau et est appelé I'EMG de surface {sEMG). Pour un muscle unique, en supposant que la relation entre l'amplitude du sEMG et la force est monotone, plusieurs études ont essayé d'estimer cette force en développant des modèles actionnés par ce signal. Toutefois, ces modèles contiennent plusieurs limites à cause des hypothèses irréalistes par rapport à l'activation neurale. Dans cette thèse, nous proposons un nouveau modèle de relation sEMG/force en intégrant ce qu'on appelle le sEMG haute définition (HD-sEMG), qui est une nouvelle technique d'enregistrement des signaux sEMG ayant démontré une meilleure estimation de la force en surmontant le problème de la position de l'électrode sur le muscle. Ce modèle de relation sEMG/force sera développé dans un contexte sans fatigue pour des contractions isométriques, isotoniques et anisotoniques du Biceps Brachii (BB) lors une flexion isométrique de l'articulation du coude à 90°
The neuromuscular and musculoskeletal systems are complex System of Systems (SoS) that perfectly interact to provide motion. This interaction is illustrated by the muscular force, generated by muscle activation driven by the Central Nervous System (CNS) which pilots joint motion. The knowledge of the force level is highly important in biomechanical and clinical applications. However, the recording of the force produced by a unique muscle is impossible using noninvasive procedures. Therefore, it is necessary to develop a way to estimate it. The muscle activation also generates another electric phenomenon, measured at the skin using electrodes, namely the surface electromyogram (sEMG). ln the biomechanics literature, several models of the sEMG/force relationship are provided. They are principally used to command musculoskeletal models. However, these models suffer from several important limitations such lacks of physiological realism, personalization, and representability when using single sEMG channel input. ln this work, we propose to construct a model of the sEMG/force relationship for the Biceps Brachii (BB) based on the data analysis of a High Density sEMG (HD-sEMG) sensor network. For this purpose, we first have to prepare the data for the processing stage by denoising the sEMG signals and removing the parasite signals. Therefore, we propose a HD-sEMG denoising procedure based on Canonical Correlation Analysis (CCA) that removes two types of noise that degrade the sEMG signals and a source separation method that combines CCA and image segmentation in order to separate the electrical activities of the BB and the Brachialis (BR). Second, we have to extract the information from an 8 X 8 HD-sEMG electrode grid in order to form the input of the sEMG/force model Thusly, we investigated different parameters that describe muscle activation and can affect the relationship shape then we applied data fusion through an image segmentation algorithm. Finally, we proposed a new HDsEMG/force relationship, using simulated data from a realistic HD-sEMG generation model of the BB and a Twitch based model to estimate a specific force profile corresponding to a specific sEMG sensor network and muscle configuration. Then, we tested this new relationship in force estimation using both machine learning and analytical approaches. This study is motivated by the impossibility of obtaining the intrinsic force from one muscle in experimentation

Projeto de Graduação apresentado à Universidade Fernando Pessoa como parte dos requisitos para obtenção do grau de Licenciada em Fisioterapia
O estudo realizado tem como objectivo descrever possíveis diferenças existentes na ativação muscular entre gêneros no músculo trapézio superior, porção descendente, com base em diferenças morfológicas e histológicas, através de EMG de superfície, no movimento de elevação no plano da escápula tendo como referência o deltoide anterior. Dezoito indivíduos, onze do sexo feminino e sete do sexo masculino, com idades compreendidas entre os 18 e 30 (26.06  4.75), participaram no estudo. Relativamente aos participantes do sexo masculino, a porção descendente do trapézio superior ativou em média 22.95ms (44.45ms) após o deltoide anterior e os indivíduos do sexo feminino ativaram-no 125.91ms (135.48ms) após o deltoide anterior. Os resultados obtidos são estatisticamente significativos, sugerindo a existência de um diferente padrão de activação muscular entre gêneros na porção descendente do trapézio superior.
The aim of the study was to describe possible differences in muscle onset between genders with sEMG, in the upper trapezius muscle, descending portion, during dynamic scapular elevation, using as reference the anterior deltoid muscle. Eighteen subjects participated in the study, seven male and eleven female. The results indicated that the male mean activation occurred sooner than their female counterparts. The onset of the upper trapezius took place 22.95ms after anterior deltoid muscle in male participants and 125.95ms on the female participants. The results were statistical significant, suggesting that a different pattern of muscle activations between genders do exist.

Despite several previous investigations, the direct correlation between the elbow joint angle and the activities of related muscles is still an unresolved topic. The sEMG signals were recorded from biceps brachii (6x8 electrodes, 10mm IED, d=3mm) and brachioradialis (1x8 electrodes, 5mm IED, d=3mm) of ten subjects. The subjects were asked to perform isometric elbow flexion at five joint angles with four contraction levels with respect to the maximum contraction (MVC) at that joint angle. The RMS values of biceps brachii (BB) and brachioradialis (BR) are computed within 500ms epoch and averaged over the muscle’s active region. These values increase along as force increases regardless the joint angle. Concerning the different joint angle, we found that as the arm extended, the RMS values of seven subjects decreased, while the RMS values of three subjects increased. This behavior suggests different strategies of muscle contribution to the task in different subjects but may also be attributed to the technical issues discussed in Chapter 2 - 7. Prior to this investigation, several issues related to the sEMG signals recording and processing were evaluated. Analysis on the effect of different elbow joint angle on the position of the innervations zone (IZ) of biceps brachii muscle indicates that the IZ shifts distally 24±9mm as the subjects extend their arms. Thus to assure sEMG signal recording, a grid of electrodes is selected instead of bipolar electrodes. The issue of spatial aliasing, which has not been addressed before, was studied. Greater electrode’s diameter implies higher spatial low pass filtering effect which gives an advantage as anti-aliasing filter in space. On the other hand, this low pass filtering effect increase the error on the power for the single sEMG image (d=10mm, 10mm IED) to 3±13.5% compared to the continuous image. Larger IED introduces RMS estimation error up to ±18% for the single sEMG image (15mm IED). However, taking the mean of a group of maps, the error of the mean is negligible (<3%). Furthermore, the envelope of the rectified EMG has been investigated. Five digital low pass filters (Butterworth, Chebyshev, Inverse Chebyshev, and Elliptic) with five different orders, four cut off frequencies and one or bi-directional filtering were tested using simulated sEMG interference signals. The results show that different filters are optimal for different applications. Power line interference is one of the sources of impurity of the sEMG signals. Notch filter, spectral interpolation, adaptive filter, and adaptive noise canceller with phase locked loop were compared. Another factor that affects the amplitude of sEMG is the subcutaneous layer thickness (ST). Higher contraction level and greater elbow joint angle lead to thinner ST. RMS values tend to decrease for thicker ST at a rate of 1.62 decade/decade.

Submitted by Erika Demachki (erikademachki@gmail.com) on 2015-01-16T17:10:19Z No. of bitstreams: 2 license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Dissertação - Gustavo Souto de Sá e Souza - 2013.pdf: 6684567 bytes, checksum: fefcf131a81d9d993c2ef7f8bf77fa9e (MD5)
Approved for entry into archive by Erika Demachki (erikademachki@gmail.com) on 2015-01-16T17:36:13Z (GMT) No. of bitstreams: 2 license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Dissertação - Gustavo Souto de Sá e Souza - 2013.pdf: 6684567 bytes, checksum: fefcf131a81d9d993c2ef7f8bf77fa9e (MD5)
Made available in DSpace on 2015-01-16T17:36:13Z (GMT). No. of bitstreams: 2 license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Dissertação - Gustavo Souto de Sá e Souza - 2013.pdf: 6684567 bytes, checksum: fefcf131a81d9d993c2ef7f8bf77fa9e (MD5) Previous issue date: 2013-12-18
This project, in the area of biomedical engineering, belongs to the promising field of research in surface electromyography (s-EMG). This technology can be used for in-depth study of some neuromuscular diseases, such as polyneuropathies and myopathies. Using an array of multichannel electrodes, we can also obtain the decomposition of s-EMG signals, estimation of conduction velocity of muscle fibers, location of innervation zones (set of motor units), among other applications. Although there are wireless electromyographers, there are no wireless electrode arrays in the market. Thinking about this, it was developed a wireless linear array of ten active electrodes for surface electromyography and a set of programs able to receive and process the data captured by this device. The hardware’s features are: low cost compared to similar equipment on the market, 12 bits resolution, 9216 samples per second (1024 samples per second per channel, with 9 channels and 10 electrodes in bipolar configuration), common mode rejection ratio greater than 50 dB; possess an interface for easy interaction with any computers via Bluetooth; enabling research in diverse areas (biomechanics, signal acquisition in athletes, animals, among other possibilities). In addition, it is powered by two lithium-ion batteries and autonomy of approximately 3 hours and 18 minutes. Although there were challenges in various stages of the device construction process, for example, in obtaining a high processing capacity and a high data transmission rate, the tests with prototypes show excellent results, consistent with the literature. After the implementation of the hardware, operational tests were performed as well as practical applications the use of a multi-channel electromyographer.
Esse projeto, da área da engenharia biomédica, pertence ao campo promissor de pesquisas em eletromiografia de superfície (EMG-s). Essa tecnologia pode ser usada para o estudo aprofundado de algumas doenças neuromusculares, como por exemplo, polineuropatias, miastenias e miopatias. Utilizando um arranjo de eletrodos multicanal, também podemos obter a decomposição de sinais de EMG-S, estimativa de velocidade de condução das fibras musculares, localização de zonas de inervação (conjunto de pontos motores), entre outras aplicações. Apesar de existirem eletromiógrafos sem fio, não há arranjos de eletrodos sem fio no mercado. Pensando nisso, foi desenvolvido um arranjo linear de dez eletrodos sem fio para eletromiografia de superfície e um conjunto de programas capazes de receber e processar os dados capturados por esse dispositivo. As características alcançadas por esse eletromiógrafo portátil são um baixo custo mesmo quando comparado aos eletromiógrafos de apenas um canal do mercado, 12 bits de resolução, 9216 amostras por segundo (1024 amostras por segundo por canal, com 9 canais e 10 eletrodos utilizando a configuração bipolar), taxa de rejeição de modo comum maior que 50 dB, uma interface que permite interação com computadores via Bluetooth, permitindo pesquisa em diversas áreas (biomecânica, aquisição de sinais em atletas, animais, entre outras possibilidades). Além disso, é alimentado por duas baterias de íon-lítio e possui uma autonomia média de 3 horas e 18 minutos. Apesar de terem surgidos desafios em várias etapas do processo de construção do dispositivo, como por exemplo, a obtenção de uma alta capacidade de processamento e de uma alta taxa de transmissão de dados, os testes com protótipos construídos mostram um resultado excelente e condizente com a literatura. Após a implementação deste hardware, foram realizados testes de funcionamento, assim como aplicações práticas da utilização de um eletromiógrafo de múltiplos canais.

L’estimation de la force générée par un muscle est importante dans les études biomécaniques et pour les applications cliniques. Puisque cette force ne peut pas être mesurée directement, le signal électromyographique de surface (SEMG), reflétant le niveau d’activation musculaire, est utilisé pour quantifier la force développée. Cependant, tous les facteurs, contrôlant une contraction isométrique, n’influencent pas la force et le SEMG simultanément. Le but de ce travail de thèse est donc de développer un modèle de simulation conjointe du SEMG et de la force, afin d’étudier la relation EMG-force. Dans ce but, nous avons d’abord développé une nouvelle méthode de simulation de la force musculaire à partir d’un modèle d’EMG existant. Le modèle complet a été testé pour le choix de la stratégie de recrutement et l’influence de la durée de la consigne. Puis, nous avons utilisé une méthode de Monte Carlo pour étudier la sensibilité du modèle aux différents paramètres physiologiques d’entrée. Deux critères existants (relations EMG-force et force-variabilité de force) ainsi qu’un nouveau critère (erreur entre la consigne de force et la force générée), ont été utilisés pour optimiser les paramètres avec une consigne de force constante. Ce nouveau critère a ensuite été utilisé avec une consigne de force variable (sinusoïdale ou triangulaire), afin d’obtenir les plages optimales des paramètres. Enfin, pour évaluer notre modèle, nous avons réalisés des expérimentations et une simulation pour le biceps. Les résultats montrent que notre modèle EMG-force est capable de simuler qualitativement les comportements réels du biceps pour les contractions isotoniques et anisotoniques
The estimation of the force generated by a muscle is important in biomechanical studies and clinical applications. As this force cannot be measured directly, the surface electromyography signal (SEMG), reflecting the level of muscle activation, is used to quantify the force developed. However, all the factors controlling an isometric contraction do not influence the force and the SEMG simultaneously. The aim of this study is to develop a simulation model of SEMG and force in order to study the EMG-force relationship. For this purpose, we first developed a new method to simulate the muscle force from an existing EMG model. We tested the complete model with two recruitment strategies and studied the influence of target force duration. Then we used a Monte Carlo method to study the sensitivity of the model to various input physiological parameters. Two existing criteria (EMG-force and force-force variability relationships) and a new criterion (error between the target force and the generated force) were used to optimize the parameters in constant target force contractions. This new criterion was then used in variable target force contractions (sinusoidal or triangular target) in order to obtain the optimum parameter ranges. Finally, to evaluate our model, we performed experiments and simulations for the biceps. The results have shown that our EMG-force model can qualitatively simulate the behaviour of the biceps for isotonic and anisotonic contractions

Motor Unit (MU) innervation zones (IZs) localization is an important step in several clinical and non-clinical applications including 1) Acquisition of sEMG signal for accurate estimation of its amplitude and other parameters by avoiding placing the electrodes on IZs, 2) Accurate estimation of the EMG-Force relationship, 3) Effective injection of Botulinum Toxin in Post-stroke Spasticity near the IZs, and 4) Guiding the obstetricians to perform episiotomy during child delivery by avoiding cutting near the IZs of External Anal Sphincter (EAS) muscle. The minimal invasive way to identify the location of the IZs generally for any muscle and specifically for EAS muscle is to use multi-channel EMG signals. MU IZs can be detected from the multi-channel sEMG signals, for a fusiform muscle if the signal is acquired with an array of electrodes placed parallel to the muscle fibers, using digital signal and image processing algorithms. As most of the signal processing algorithms work on an adequate quality of the signal, thus before detecting the innervation zone it is made sure that the signal is of good quality. For this purpose, a method based on statistical thresholding of various parameters is proposed to detect the bad channels in the sEMG signals. If the number of the bad consecutive channels are more than 2 then it is suggested to acquire the signal again, otherwise each bad channel is approximated by the interpolation of its neighbor channels. As some background noise is always acquired with the EMG signal so further image enhancement techniques are used to enhance the MUAP propagation region in the spatio-temporal images and suppress the background noise. The MUAP pattern is then detected in the spatio-temporal sEMG images using multi-scale Hessian based filtering and the corresponding MU IZs are identified as the starting point of propagation of the MUAP. A software is also developed which can be used to visualize the signals acquired from EAS, detect and display the IZs and more importantly compute and display the histogram of the IZs and generate reports which will help the obstetrician while performing episiotomy during child delivery to avoid cutting vulnerable regions that may lead to fecal incontinence at later age.

En raison de leur implication potentielle dans la performance et la blessure, la compréhension des coordinations musculaires des ischio-jambiers représente aujourd’hui un enjeu en sciences du mouvement. Cependant, les méthodes actuelles basées sur la mesure de l’activation musculaire ou la modélisation musculo-squelettique ne permettent pas d’identifier les facteurs influençant les coordinations pour chaque individu. Explorer ces facteurs permettrait de mieux comprendre comment le système nerveux central explore et exploite l’ensemble des coordinations musculaires possibles afin d’adopter une solution permettant de réaliser la tâche demandée. Ce manuscrit a pour objectif de décrire les coordinations musculaires des ischio-jambiers à partir de données acquises in vivo. La partie expérimentale de ce travail s’articule autour de trois études qui ont exploré le couplage entre l’activation musculaire et les propriétés mécaniques des muscles. Ces travaux ont visé à comprendre les facteurs déterminants les coordinations de chaque individu, l’effet des coordinations musculaires de chaque individu sur leur performance et leur adaptation après une blessure musculaire.Les résultats de nos études montrent que les coordinations musculaires des ischio-jambiers varient substantiellement parmi les personnes actives et les athlètes de haut niveau. Nous avons montré que ces coordinations ne sont pas déterminées par les propriétés mécaniques de chaque muscle, i.e. le déséquilibre d’activation entre les ischio-jambiers n’est pas lié au déséquilibre de capacité de production de force. Ces coordinations apparaissent également plus ou moins avantageuses pour être performant dans une tâche d’endurance réalisée jusqu’à épuisement. Enfin, après la survenue d’une blessure, les coordinations musculaires diffèrent entre les jambes blessée et non blessée. Plus précisément, la contribution du muscle blessé au couple de force total est plus faible en comparaison du même muscle de la jambe opposée. Ces différences pourraient avoir des conséquences négatives à court-terme (pour la performance) et à long terme, pouvant hypothétiquement aller jusqu’à la récidive de la blessure
The understanding of hamstrings coordination is a trending topic in human movement science because of their potential involvement in both performance and injury. However, current experimental methods based on muscle activation recording or musculoskeletal modeling do not enable the identification of the factors that influencing muscle coordination for each individual. Addressing this issue is fundamental to understand how the central nervous system explores and exploits a set of many feasible coordination to adopt a good enough solution that enables producing a given motor task. This manuscript aims at describing hamstring coordination with in vivo data. The experimental part of this work is based on three studies that explored the coupling between muscle activation and the hamstrings mechanical properties. Also, this work aims at understanding the factors that influence muscle coordination of each individual, the effect of muscle coordination on their performance and their adaptation after a muscle injury.Our results show that hamstrings coordination varies substantially among active and elite athletes. We have shown that hamstrings coordination was not determined by the mechanical properties of each muscle, i.e. the imbalance of activation between hamstrings is not related to the imbalance of force production capacity. In addition, hamstrings coordination appears more or less advantageous in order to perform during a time to exhaustion task. Finally, muscle coordination differs between injured and uninjured legs, even after the completion of rehabilitation. Specifically, the injured muscle contributes in a lower extent to hamstring total torque in comparison with the same muscle of the opposite leg. These differences could have negative consequences in the short term (for performance) and in the long term, which could hypothetically increase the risk of reinjury

Projeto de Graduação apresentado à Universidade Fernando Pessoa como parte dos requisitos para obtenção do grau de Licenciada em Fisioterapia
Resumo: Objetivo: Analisar a atividade eletromiográfica dos músculos serrátil anterior, trapézio superior e inferior ao longo do exercício de box position e prancha com diferentes variações e superfícies. Metodologia: Foi realizada a análise da atividade electromiográfica dos músculos acima referidos em quinze participantes (média de idades 22.00±2.00 anos; IMC 23.41±3.83 kg/m2) nas diferentes variações dos exercícios box position e prancha, sendo os elétrodos colocados no membro superior dominante dos participantes. Resultados: Foram encontradas diferenças significativas quando analisada a atividade electromiográfica nos diversos músculos em cada variação mas também em cada músculo entre as variações. O músculo serrátil anterior foi o que evidenciou uma maior atividade muscular, com cerca de 38,41% de contração máxima voluntária, sendo que os valores mais altos foram observados em superfícies estáveis. O músculo que evidenciou menos atividade muscular foi o trapézio superior. Conclusão: A atividade electromiográfica dos músculos em questão variam consoante o tipo de variação dos exercícios efetuados. No entanto o músculo serrátil anterior apresentou os maiores níveis de atividade eletromiográfica em todas as variações dos exercícios propostos.
Objective: The aim of the present study is to analyze the electromyographic activity of the anterior serratus muscle, upper and lower trapezius throughout the box position and plank exercise, with different variations and surfaces. Methodology: The electromyographic activity of announced muscles was realized in fifteen participants (mean age 22.00±2.00 years; BMI 23.41±3.83 kg/m2) in different variations of box position and plank, being placed electrodes according to the preference of the participants. Results: Significant differences were found when the electromyographic activity were analyzed in the various muscles in each variation but also in every muscle between the variations. The anterior serratus muscle was what showed greater muscular activity, with value of 38.41% of maximum voluntary contraction, with the higher values verified on stable surfaces. The muscle that showed less muscular activity was the upper trapezius muscle. Conclusion: The present electromyographic muscular activity changes depending on the type of variation of performed exercises. However, the anterior serratus muscle presented higher levels of electromyographic activity in each exercise variation proposed.
N/A

Projeto de Graduação apresentado à Universidade Fernando Pessoa como parte dos requisitos para obtenção do grau de Licenciado em Fisioterapia
Objetivo: O presente estudo tem como objetivo avaliar a atividade neuromuscular dos músculos peitoral, serrátil anterior, infra-espinhoso e grande dorsal durante o exercício de press up box, com diferentes variações. Metodologia: A atividade electromiográfica dos músculos selecionados de doze participantes do sexo masculino (Média de idades 25,08±4,26 anos; IMC 22,66±1,85 Kg/m2) foi analisada durante diferentes variações do exercício press up box. Resultados: Foram encontradas diferenças significativas na atividade electromiográfica dos diferentes músculos em cada variação e na atividade electromiográfica de cada músculo nas diferentes variações. O músculo que apresentou maior ativação foi o grande peitoral, porção esternal, com um valor médio de 22% de contração máxima voluntária, sendo a maior contribuição em superfícies instáveis. Já em superfície estável os músculos serrátil anterior e infra-espinhoso atingiram valores de 13 e 15% de CMV. O músculo que apresentou menor atividade muscular foi o grande dorsal. Conclusão: A atividade electromiográfica dos músculos estudados foi dependente do tipo de variação do exercício press up box imposta.
Objective: The aim of the present study is to describe the electromyographic activity (sEMG) of the serratus anterior, pectoralis sternal and clavicular portion, infra-spinous and latissimus dorsi muscles, during the press-up box (PUB) exercise, with different variations. Methodology: The sEMG of selected muscles from twelve male participants (Mean age 25.08 ± 4.26 years; BMI 22.66 ± 1.85 kg / m2) were analyzed for different variations of the press up box. Results: Statistically significant differences were found for sEMG between different muscles in each variation, and for each muscle across different variations. Data analysis also suggests that the pectoralis major muscle, sternal portion, presented the greatest activation (22% of the maximum voluntary contraction) mostly on unstable conditions. Moreover, with stable surface the serratus anterior and the infra-spinous muscles reached the values of 13 and 15% of MVC. The muscle that presented the lowest muscle activity was the latissimus dorsi. Conclusion: The sEMG of the studied muscles was dependent on the type of variation of the press up box exercise.

Les signaux électromyographiques de surface (sEMG) correspondent aux signaux électriques composés par les potentiels d’action produits par les unités motrices d’un muscle actif et enregistrés par des électrodes de surface. Les signaux sEMG sont largement utilisés dans la médicine, le contrôle des prothèses et plus généralement dans les études biomécaniques portant sur l’analyse du mouvement humain. Les signaux sEMG sont très souvent utilisés comme un indicateur d’activation musculaire.Bien que présentant un intérêt évident, l’utilisation de ces signaux reste difficile compte tenu qu’ils sont souvent susceptibles d’interférence (diaphonie, ou plus communément « crosstalk ») entre les muscles contigus, parfois même éloignés. Cette contamination croisée est particulièrement présente pour des muscles présents dans un volume restreint, ce qui est le cas des muscles extenseur de l’index et du petit doigt, extensor indicis et extensor digiti minimi. L’interférence induit la réduction de la précision de l’estimation des activations musculaires et reste, à ce titre, un problème important et récurrent de la biomécanique. Afin que les signaux sEMG puissent être utilisés de manière plus robuste en biomécanique, il convient de réduire cette interférence avant de procéder à l’estimation des activations musculaires. Les activations individuelles des muscles participant au mouvement correctement estimées peuvent être utilisées comme données d’entrées d’un modèle biomécanique. Cette démarche, nommée dynamique directe, permet notamment d’estimer la force externe produite par le système et dans un second temps de comparer cette dernière avec la mesure réalisée grâce à un système dynamométrique. En ce sens cette démarche permet une validation indirecte des estimations réalisées à partir des signaux sEMG. Dans le cadre de cette thèse, nous avons modélisé le doigt et plus particulièrement le mécanisme extenseur qui est une structure qui transmet les forces des muscles-extenseurs aux articulations digitales. Cette structure est très mal connue du point de vue biomécanique et le plus souvent représentée par un ensemble des coefficients établis sur l’analyse de mains de cadavres dans des situations très particulières et standardisées (doigts en extension). Ainsi, l’objectif de ce travail de thèse était double : (1) améliorer l’estimation de la force au bout du doigt à partir des mélanges des sEMG sur la base d’extraction des activations des signaux sEMG des muscles extensor indicis et extensor digiti minimi, et (2) modélisation biomécanique du mécanisme extenseur du doigt. Pour cela, les signaux sEMG ont été enregistrés avec une matrice d’électrodes de surface haute densité à 64 capteurs. Ensuite, l’extraction des activations musculaires a été réalisée sur la base d’une procédure de classification des potentiels détectés en utilisant les invariants musculaires que sont la direction de propagation et la profondeur de l’unité motrice à l’origine du signal.Dans un deuxième temps, un modèle biomécanique précis du mécanisme extenseur du doigt a été créé, qui contient les tendons et les principaux ligaments représentés par des bandes et des surfaces élastiques. Un algorithme de paramétrage du modèle a été proposé. Ce type d ‘approche est nécessaire pour mieux décrire les déformations du système anatomique dans des situations de mouvement sain ou pathologique.Cette démarche a montré qu’elle était pertinente pour l’étude biomécanique du doigt. Elle présente des utilisations judicieuses pour les études biomécaniques portant sur l’évaluation clinique, la réhabilitation et le contrôle des prothèses myoélectriques
The surface electromyographic signals (SEMG) are the electric signals, composed of electric potentials. These potentials are produced by the recruited motor units of an active muscle and captured by the surface electrodes. The SEMG signals are widely used in medicine, prosthesis control and biomechanical studies as an indicator of muscle activity.However, SEMG measurements are usually subjects of crosstalk or interference from nearby muscles. It appears when two or more muscles situated close to each other are active during a SEMG recording. An example of such muscles are the extensors of index and little finger, extensor indicis and extensor digiti minimi, situated close to each other and creating a significant amount of mutual crosstalk when simultaneously active. The crosstalk causes precision decrease of SEMG-based estimation of muscle activations. Hence, the crosstalk-reducing problem must be preliminary solved before muscle activation evaluation.Once the activations of individual muscles are estimated from the mixture, they may be used as an input of a finger biomechanical model to calculate a fingertip force. These models usually contain an extensor mechanism of the finger, which is a structure, transmitting the force from the extensor muscles to the finger joints. This structure is often taken into account as a set of coefficients. However, there is a lack of study about how these coefficients vary with posture, applied force, and subject variability.The purpose of this work is to improve the finger force estimation from the crosstalk-contaminated signals for isometric tasks by extracting the activations of individual muscles and improving the finger biomechanical model.Firstly, the SEMG signals were recorded with high-density surface electromyographic (HD-EMG) electrode matrix. The extraction was based on classifying the detected potentials according their propagation direction and depth of originating motor unit.Secondly, a precise biomechanical model of the finger extensor mechanism was created, containing the principal tendons and ligaments. The algorithm of the model parametrization was proposed as well.The proposed methods of muscle activation estimation along with the created extensor mechanism model may be used for calculating the fingertip force and internal tissues deformations for normal or pathological fingers

Les systèmes neuromusculaire et musculosquelettique sont des systèmes de systèmes complexes qui interagissent parfaitement entre eux afin de produire le mouvement. En y regardant de plus près, ce mouvement est la résultante d'une force musculaire créée à partir d'une activation du muscle par le système nerveux central. En parallèle de cette activité mécanique, le muscle produit aussi une activité électrique elle aussi contrôlée par la même activation. Cette activité électrique peut être mesurée à la surface de la peau à l'aide d'électrode, ce signal enregistré par l'électrode se nomme le signal Électromyogramme de surface (sEMG). Comprendre comment ces résultats de l'activation du muscle sont générés est primordial en biomécanique ou pour des applications cliniques. Évaluer et quantifier ces interactions intervenant durant la contraction musculaire est difficile et complexe à étudier dans des conditions expérimentales. Par conséquent, il est nécessaire de développer un moyen pour pouvoir décrire et estimer ces interactions. Dans la littérature de la bioingénierie, plusieurs modèles de génération de signaux sEMG et de force ont été publiés. Ces modèles sont principalement utilisés pour décrire une partie des résultats de la contraction musculaire. Ces modèles souffrent de plusieurs limites telles que le manque de réalisme physiologique, la personnalisation des paramètres, ou la représentativité lorsqu'un muscle complet est considéré. Dans ce travail de thèse, nous nous proposons de développer un modèle biofidèle, personnalisable et rapide décrivant l'activité électrique et mécanique du muscle en contraction isométrique. Pour se faire, nous proposons d'abord un modèle décrivant l'activité électrique du muscle à la surface de la peau. Cette activité électrique sera commandé par une commande volontaire venant du système nerveux périphérique, qui va activer les fibres musculaires qui vont alors dépolariser leur membrane. Cette dépolarisation sera alors filtrée par le volume conducteur afin d'obtenir l'activité électrique à la surface de la peau. Une fois cette activité obtenue, le système d'enregistrement décrivant une grille d'électrode à haute densité (HD-sEMG) est modélisée à la surface de la peau afin d'obtenir les signaux sEMG à partir d'une intégration surfacique sous le domaine de l'électrode. Dans ce modèle de génération de l'activité électrique, le membre est considéré cylindrique et multi couches avec la considération des tissus musculaire, adipeux et la peau. Par la suite, nous proposons un modèle mécanique du muscle décrit à l'échelle de l'Unité Motrice (UM). L'ensemble des résultats mécaniques de la contraction musculaire (force, raideur et déformation) sont déterminées à partir de la même commande excitatrice du système nerveux périphérique. Ce modèle est basé sur le modèle de coulissement des filaments d'actine-myosine proposé par Huxley que l'on modélise à l'échelle UM en utilisant la théorie des moments utilisée par Zahalak. Ce modèle mécanique est validé avec un profil de force enregistré sur un sujet paraplégique avec un implant de stimulation neurale. Finalement, nous proposons aussi trois applications des modèles proposés afin d'illustrer leurs fiabilités ainsi que leurs utilité. Tout d'abord une analyse de sensibilité globale des paramètres de la grille HDsEMG est présentée. Puis, nous présenterons un travail fait en collaboration avec une autre doctorante une nouvelle étude plus précise sur la modélisation de la relation HDsEMG/force en personnalisant les paramètres afin de mimer au mieux le comportement du Biceps Brachii. Pour conclure, nous proposons un dernier modèle quasi­ dynamique décrivant l'activité électro-mécanique du muscle en contraction isométrique. Ce modèle déformable va actualiser l'anatomie cylindrique du membre sous une hypothèse isovolumique du muscle
The neuromuscular and musculoskeletal systems are complex System of Systems (SoS) that perfectly interact to provide motion. From this interaction, muscular force is generated from the muscle activation commanded by the Central Nervous System (CNS) that pilots joint motion. In parallel an electrical activity of the muscle is generated driven by the same command of the CNS. This electrical activity can be measured at the skin surface using electrodes, namely the surface electromyogram (sEMG). The knowledge of how these muscle out comes are generated is highly important in biomechanical and clinical applications. Evaluating and quantifying the interactions arising during the muscle activation are hard and complex to investigate in experimental conditions. Therefore, it is necessary to develop a way to describe and estimate it. In the bioengineering literature, several models of the sEMG and the force generation are provided. They are principally used to describe subparts of themuscular outcomes. These models suffer from several important limitations such lacks of physiological realism, personalization, and representability when a complete muscle is considered. In this work, we propose to construct bioreliable, personalized and fast models describing electrical and mechanical activities of the muscle during contraction. For this purpose, we first propose a model describing the electrical activity at the skin surface of the muscle where this electrical activity is determined from a voluntary command of the Peripheral Nervous System (PNS), activating the muscle fibers that generate a depolarization of their membrane that is filtered by the limbvolume. Once this electrical activity is computed, the recording system, i.e. the High Density sEMG (HD-sEMG) grid is define over the skin where the sEMG signal is determined as a numerical integration of the electrical activity under the electrode area. In this model, the limb is considered as a multilayered cylinder where muscle, adipose and skin tissues are described. Therefore, we propose a mechanical model described at the Motor Unit (MU) scale. The mechanical outcomes (muscle force, stiffness and deformation) are determined from the same voluntary command of the PNS, and is based on the Huxley sliding filaments model upscale at the MU scale using the distribution-moment theory proposed by Zahalak. This model is validated with force profile recorded from a subject implanted with an electrical stimulation device. Finally, we proposed three applications of the proposed models to illustrate their reliability and usefulness. A global sensitivity analysis of the statistics computed over the sEMG signals according to variation of the HD-sEMG electrode grid is performed. Then, we proposed in collaboration a new HDsEMG/force relationship, using personalized simulated data of the Biceps Brachii from the electrical model and a Twitch based model to estimate a specific force profile corresponding to a specific sEMG sensor network and muscle configuration. To conclude, a deformableelectro-mechanicalmodelcouplingthetwoproposedmodelsisproposed. This deformable model updates the limb cylinder anatomy considering isovolumic assumption and respecting incompressible property of the muscle

L'objectif de ce travail de thèse est d'étudier d'une part les spécificités de la commande nerveuse lors de la contraction excentrique en explorant les mécanismes impliqués au niveau spinal et d'autre part d'examiner les mécanismes nerveux responsables de la plasticité du système neuromusculaire après un entraînement de force excentrique sous-maximal. A travers ce travail de thèse, nous mettons en évidence la contribution de l'inhibition récurrente à la réduction de l'activation musculaire classiquement observée lors de la contraction excentrique. Par ailleurs, nous montrons que l'inhibition récurrente est majorée lors des contractions sous-maximales indépendamment du mode de contraction. Ces résultats soulignent le rôle important de l'inhibition récurrente dans la spécificité de la commande nerveuse lors de la contraction excentrique. Nous confirmons que le pilotage nerveux de la contraction excentrique peut être modulé par l'entraînement de force excentrique même si les modulations de l'excitabilité spinale semblent dépendre des caractéristiques de l'entraînement
The purpose of this PhD research is, on the one hand, to study the neural drive specificities during eccentric contractions by exploring the neural mechanisms involved at spinal level and, on the other hand, to examine the neural mechanisms responsible for the modulations of neuromuscular system following a strength submaximal eccentric training. Through this PhD research we highlight the contribution of recurrent inhibition by the Renshaw cell to the decrease of muscular activation typically observed during eccentric contraction. Furthermore, we show that recurrent inhibition is enhanced during submaximal contractions regardless of the contraction type. These results emphasize the important role of recurrent inhibition in the specificity of neural control during eccentric contractions. We confirm that the neural drive of the eccentric contraction may be modulated by eccentric strength training although modulations of spinal excitability seem to depend on the characteristics of training

碩士
國立高雄師範大學
聽力學與語言治療研究所
101
surface electromyography(sEMG) during swallowing from stroke patients with Dysphagia Abstract Dysphagia is a common complication in stroke patient. It not only impedes the quality of life but also increases the risk of pulmonary complications and even mortality. The videofluoroscopic swallowing study are take as the golden standard methods to assess dysphagia. However, it can’t be performed in the bedside. Our purpose is to investigate whether there is a difference of the sEMG during swallowing between normal population and stroke patients with dysphagia. After analyzing the signals, sEMG may be used as a quantifiable tools for dysphagia evaluation over the bedside. We obtained sEMG during swallowing, which consist of bilateral swallowing myoelectric signals, and compared the difference between stroke patients with dysphagia and normal population. We follow the method of “Vaiman(2007) sEMG swallowing evaluation process” when designing our study project. We recruited 20 stroke patients with dysphagia , and 20 normal subjects. Of all the participates, sEMG of four group of muscles(both sides) including obicularis oris,masseter,submental muscles and laryngeal strap muscles,during swallowing of 5 c.c. of water were recorded, Of the recorded sEMG, 7 variables such as baseline, average amplitude, peak amplitude, duration, peak latency, onset and offset relative to the orbicularis oris were analyzed. Independent t test were used to assess the inter-group difference. Results are as followed. 1. In stroke group, difference between sound side and hemi-side are significantly greater than those of normal group. The significant different variables contains: (1)Baseline, average amplitude and peak latency of orbicularis oris。 (2)Onset time of masseter and submental muscles groups。 (3)Average amplitude, peak amplitude and duration of laryngeal strap muscles。 2.When comparing the sound side and hemi-side of the stroke group, we found that except for the baseline of orbicularis oris at the sound side is higher than the hemi-side, there is no significant difference among the other parameters. Whereas, we can still see some trend of these parameters as followed. ＊The average amplitude of orbicularis oris, masseter, submental muscles group at the sound side are higher than hemi-side. Also, the duration of the sound side is longer than hemi-side. ＊The average amplitude and peak amplitude of laryngeal strap muscle group of the hemi-side is higher than sound side. Also, the duration of hemi-side is longer than the sound side. 3.When evaluating the relevant coefficient of all parameters and functional oral intake scale(FOIS), We found that only the average amplitude and peak amplitude of the masseter is significantly related with FOIS. Therefore, we concluded that sEMG recorded can only reflect how the swallowing muscles contract but still can’t be use to measure or interpret one’s functional oral intake ability. 28 swallowing electromyographic parameters have been analyzed and only 8 out of 28 (26％) shows significant difference. Among the parameters during pharyngeal phase of swallowing, the onset time of the submental muscles group in patient group is significantly greater than the normal group. Also, due to possible compensatory effects of laryngeal strap muscles group in stroke patients, it results in stronger power and longer contraction time over the hemi-side muscles than sound side. Eventually,it leads to the reason why the difference between hemi-side and sound-side in patient group is significantly smaller than the normal group. Due to the possible compensatory effect developed in stroke patient, the usage and interpretation of sEMG in assessing dysphagia. becomes too complicated and might be misleading. We concluded that in current acknowledge, it is not suitable to use sEMG for the evaluation of stroke patients with dysphagia. Whereas, the model of our study can still be further used to similar studies for different types of patients. key words: stroke、dysphagia、surface elecyromyography。

碩士
國立臺灣大學
資訊工程學研究所
107
Strength training improves overall health, well-being, physical appearance, and sports performance.There are four major factors that affect training efficacy in a training session: exercise type, number of repetitions, movement velocity, and workload. Prior research has used wearable sensors to detect exercise type, number of repetitions, and movement velocity while training. However, detecting workload still requires instrumentation of exercise equipment such as exercise machines, or free weights. This paper presents MuscleSense, an approach that detects training weight through wearable devices. In particular, MuscleSense uses various regressors to predicting weight using signals from wearable sEMG sensors mounted on user''s arm or forearm. We evaluated the effects of sensor placement and collected training data from 20 participants. The results from our user study show that MuscleSense achieves Root Mean Square Error(RMSE) of 0.683kg in sensing workload through sensors data from both forearm and arm using Support Vector Regressor of linear kernel.

This thesis presents an activity mode intent recognition approach for safe, robust and reliable control of powered backbone exoskeleton. The thesis presents the background and a concept for a powered backbone exoskeleton that would work in parallel with a user. The necessary prerequisites for the thesis are presented, including the collection and processing of surface electromyography signals and inertial sensor data to recognize the user’s activity. The development of activity mode intent recognizer was described based on decision tree classification in order to leverage its computational efficiency. The intent recognizer is a high-level supervisory controller that belongs to a three-level control structure for a powered backbone exoskeleton. The recognizer uses surface electromyography and inertial signals as the input and CART (classification and regression tree) as the classifier. The experimental results indicate that the recognizer can extract the user’s intent with minimal delay. The approach achieves a low recognition error rate and a user-unperceived latency by using sliding overlapped analysis window. The approach shows great potential for future implementation on a prototype backbone exoskeleton.

Modelling of human motion is used in a wide range of applications. An important aspect of accurate representation of human movement is the ability to customize models to account for individual differences. The following work proposes a methodology using Hill-based candidate functions in the Fast Orthogonal Search (FOS) method to predict translational force at the wrist from flexion and extension torque at the elbow. Within this force estimation framework, it is possible to implicitly estimate subject-specific physiological parameters of Hill-based models of upper arm muscles. Surface EMG data from three muscles of the upper arm (biceps brachii, brachioradialis and triceps brachii) were recorded from 10 subjects as they performed isometric contractions at varying elbow joint angles. Estimated muscle activation level and joint kinematic data (joint angle and angular velocity) were utilized as inputs to the FOS model. The resulting wrist force estimations were found to be more accurate for models utilizing Hill-based candidate functions, than models utilizing candidate functions that were not physiologically relevant. Subject-specific estimates of optimal joint angle were determined via frequency analysis of the selected FOS candidate functions. Subject-specific optimal joint angle estimates demonstrated low variability and fell within the range of angles presented in the literature.
Thesis (Master, Electrical & Computer Engineering) -- Queen's University, 2008-10-30 01:32:01.606

Title: Strategy of postural stabilisation using unstable surface and water barrel. Objectives: The aim of this study is to investigate the level of muscle actiavtion of choosen muscles during lunge on unstable surface or with using water barrel. Investigation of the postural strategy used during lunge orn unstable surface or with the water barrel and creation of methodology for measurment and data analysis. Methods: Into this pilot study, there were picked 5 people (athletes). Data for outcomes where used from 3 athletes. The measurement of level of muscle activation were done by surface electromyography over gluteus medius muscle and musculi multifidii bilaterally.For data procession was used software Origin 2012 Postural stability was measured through force plates by Kistler and gained data were procesed by using software programmes Bioware, MS Excel and Matlab. For data analysis from EMG measurement was used simple comparasion of outcomes. Stabilometry outcomes were analysed by statistical methode t-test. Results: The results indicate greater level of muscle activation during lunge with aquabag than lunge on unstable surface for most of the measured muscles. The only exception, where level of muscle activity was higher, was multifidi muscle on rear leg. Another outcome proved that the postural...

Title: Strategy of postural stabilisation using unstable surface and water barrel. Objectives: The aim of this study is to investigate the level of muscle actiavtion of choosen muscles during lunge on unstable surface or with using water barrel. Investigation of the postural strategy used during lunge orn unstable surface or with the water barrel and creation of methodology for measurment and data analysis. Methods: Into this pilot study, there were picked 5 people (athletes). Data for outcomes where used from 3 athletes. The measurement of level of muscle activation were done by surface electromyography over gluteus medius muscle and musculi multifidii bilaterally.For data procession was used software Origin 2012 Postural stability was measured through force plates by Kistler and gained data were procesed by using software programmes Bioware, MS Excel and Matlab. For data analysis from EMG measurement was used simple comparasion of outcomes. Stabilometry outcomes were analysed by statistical methode t-test. Results: The results indicate greater level of measured trunk muscles activation during lunge with aquabag than lunge on unstable surface. Another thing which was found is that there was higher activation of Gluteus medius muscle on dominant lower extremity when performin lunge on unstable...

The Master's thesis is focused on monitoring of the core muscles and m. biceps femoris activity in postural stability examination by Dynamic Neuromuscular Stabilization tests using surface electromyography in patients with LBP. Trunk extension test, trunk flexion test and squat test were chosen. The aim of this study was comparing muscle activity timing, symmetry of muscle activation and average activation between healthy subjects and patients with LBP. The changes of muscle activation pattern were observed and average activation was determined. In theoretical part the principles of DNS, surface electromyography and the relationship between postural stabilization and LBP are discussed. Two groups were measured - patients with LBP and a control group, 20 subjects together. The onset times were measured and rank of activation for each test and for each subject was determined. Then the average order for whole group was specified. Measured data were evaluated according to Wilcoxon signed-rank test. In average activation there was analysed change of activation in time. Results were compared between two measured groups. There is significant difference between the two groups in the rank of activation of m. ES l. dx. in trunk extension test and m. EO in squat test. Assymetry in timing of the right and left...

In the field of ergonomics, biomechanics, sports and rehabilitation muscle fatigue is regarded as an important aspect since muscle fatigue is considered to be one of the main reasons for musculoskeletal disorders. Classical signal processing techniques used to understand muscle behavior are mainly based on spectral based parameters estimation, and mostly applied during static contraction and the signal must be stationary within the analysis window; otherwise, the resulting spectrum will make little physical sense. Furthermore, the shape and size of the analysis window also directly affect the spectral estimation. But fatigue analysis in dynamic conditions is of utmost requirement because of its daily life applicability. It is really difficult to consistently find the muscle fatigue during dynamic contraction due to the inherent non-stationary nature and associated noise in the signal along with complex physiological changes in muscles. Nowadays, in addition to linear signal processing, different non-linear signal processing techniques are adopted to find out the consistent and robust indicator for muscle fatigue under dynamic condition considering the high degree of non-linearity (caused by functional interference between different muscles, changes of signal sources and paths to recording electrodes, variable electrode interface etc.) in the signal. In this work, various linear and nonlinear-non-stationary signal processing methods, applied on surface EMG signal for muscular fatigue analysis under dynamic contraction are studied. In present study, surface EMG (sEMG) signals are recorded from Biceps Brachii muscles from eight (N=8) physically active college students during dynamic lifting 7 kg load at the rate of 20 lifts/min till they become fatigue. EMG data is processed in two ways -1. taking the whole EMG response and 2. breaking into three ranges of contraction (0-45)o, (45-90)o and >90o, to study better response region. It is observed that in spectral estimation techniques auto-regressive (AR) based spectral estimation technique gives better frequency resolution than periodogram for small epochs, as AR is based on parametric estimation. Both the previous methods provide only the frequency information in the signal. In order to estimate the time varying nature of frequency content in a signal various time-frequency signal processing techniques are used like – Short Time-Fourier Transform (STFT), Smoothed pseudo Wigner-Ville (SPWD), Choi-William distribution (CWD), Continuous Wavelet Transform (CWT), Huang-Hilbert Transform (HHT) and Recurrence Quantification Analysis (RQA) are used. The last two techniques are used by considering the EMG signal as non-linear and non-stationary signals. Among these techniques, STFT is the simplest time-frequency analysis technique. But tradeoff between time and frequency resolution is the major constraint in STFT, therefore, a window length of 256 samples are considered in this study. In order to tackle time-frequency resolution problem different Cohen-class distribution techniques are used like SPWD and CWD, where the result is severely affected by the presence of interference terms which make its interpretation really difficult. Different adaptive filters are used in SPWD and CWD to suppress these interference terms during analysis. Among these time-frequency analysis techniques continuous wavelet transform provides the most accurate results in comparison to other time-frequency analysis techniques. Similar result is obtained in present study. This fatigue response is further improved using non-linear and non-stationary techniques like HHT and RQA. HHT shows less variation in frequency response than CWT analysis result. Percentage of determinism calculated using recurrence quantification analysis method is found to be more sensitive than mean frequency estimation. Therefore, non-linear and non-stationary signal processing techniques are to be better indicator of muscle fatigue during dynamic contraction.

In the field of ergonomics, biomechanics, sports and rehabilitation muscle fatigue is regarded as an important aspect since muscle fatigue is considered to be one of the main reasons for musculoskeletal disorders. Classical signal processing techniques used to understand muscle behavior are mainly based on spectral based parameters estimation, and mostly applied during static contraction and the signal must be stationary within the analysis window; otherwise, the resulting spectrum will make little physical sense. Furthermore, the shape and size of the analysis window also directly affect the spectral estimation. But fatigue analysis in dynamic conditions is of utmost requirement because of its daily life applicability. It is really difficult to consistently find the muscle fatigue during dynamic contraction due to the inherent non-stationary nature and associated noise in the signal along with complex physiological changes in muscles. Nowadays, in addition to linear signal processing, different non-linear signal processing techniques are adopted to find out the consistent and robust indicator for muscle fatigue under dynamic condition considering the high degree of non-linearity (caused by functional interference between different muscles, changes of signal sources and paths to recording electrodes, variable electrode interface etc.) in the signal. In this work, various linear and nonlinear-non-stationary signal processing methods, applied on surface EMG signal for muscular fatigue analysis under dynamic contraction are studied. In present study, surface EMG (sEMG) signals are recorded from Biceps Brachii muscles from eight (N=8) physically active college students during dynamic lifting 7 kg load at the rate of 20 lifts/min till they become fatigue. EMG data is processed in two ways -1. taking the whole EMG response and 2. breaking into three ranges of contraction (0-45)o, (45-90)o and >90o, to study better response region. It is observed that in spectral estimation techniques auto-regressive (AR) based spectral estimation technique gives better frequency resolution than periodogram for small epochs, as AR is based on parametric estimation. Both the previous methods provide only the frequency information in the signal. In order to estimate the time varying nature of frequency content in a signal various time-frequency signal processing techniques are used like – Short Time-Fourier Transform (STFT), Smoothed pseudo Wigner-Ville (SPWD), Choi-William distribution (CWD), Continuous Wavelet Transform (CWT), Huang-Hilbert Transform (HHT) and Recurrence Quantification Analysis (RQA) are used. The last two techniques are used by considering the EMG signal as non-linear and non-stationary signals. Among these techniques, STFT is the simplest time-frequency analysis technique. But tradeoff between time and frequency resolution is the major constraint in STFT, therefore, a window length of 256 samples are considered in this study. In order to tackle time-frequency resolution problem different Cohen-class distribution techniques are used like SPWD and CWD, where the result is severely affected by the presence of interference terms which make its interpretation really difficult. Different adaptive filters are used in SPWD and CWD to suppress these interference terms during analysis. Among these time-frequency analysis techniques continuous wavelet transform provides the most accurate results in comparison to other time-frequency analysis techniques. Similar result is obtained in present study. This fatigue response is further improved using non-linear and non-stationary techniques like HHT and RQA. HHT shows less variation in frequency response than CWT analysis result. Percentage of determinism calculated using recurrence quantification analysis method is found to be more sensitive than mean frequency estimation. Therefore, non-linear and non-stationary signal processing techniques are to be better indicator of muscle fatigue during dynamic contraction.

A finite element (FE) model for the generation of single fiber action potentials (SFAPs) in a muscle undergoing various degrees of fiber shortening has been developed. The muscle is assumed to be fusiform with muscle fibers following a curvilinear path described by a Gaussian function. Different degrees of fiber shortening are simulated by changing the parameters of the fiber path and maintaining the volume of the muscle constant. The conductivity tensor is adapted to the muscle fiber orientation. At each point of the volume conductor, the conductivity of the muscle tissue in the direction of the fiber is larger than that in the transversal direction. Thus, the conductivity tensor changes point-by-point with fiber shortening, adapting to the fiber paths. An analytical derivation of the conductivity tensor is provided. The volume conductor is then studied with an FE approach using the analytically derived conductivity tensor (Mesin, Joubert, Hanekom, Merletti&Farina 2006). Representative simulations of SFAPs with the muscle at different degrees of shortening are presented. It is shown that the geometrical changes in the muscle, which imply changes in the conductivity tensor, determine important variations in action potential shape, thus affecting its amplitude and frequency content. The model is expanded to include the simulation of motor unit action potentials (MUAPs). Expanding the model was done by assigning each single fiber (SF) in the motor unit (MU) a random starting position chosen from a normal distribution. For the model 300 SFs are included in an MU, with an innervation zone spread of 12 mm. Only spatial distribution was implemented. Conduction velocity (CV) was the same for all fibers of the MU. Representative simulations for the MUAPs with the muscle at different degrees of shortening are presented. The influence of interelectrode distance and angular displacement are also investigated as well as the influence of the inclusion of the conductivity tensor. It has been found that the interpretation of surface electromyography during movement or joint angle change is complicated owing to geometrical artefacts i.e. the shift of the electrodes relative to the muscle fibers and also because of the changes in the conductive properties of the tissue separating the electrode from the muscle fibers. Detection systems and electrode placement should be chosen with care. The model provides a new tool for interpreting surface electromyography (sEMG) signal features with changes in muscle geometry, as happens during dynamic contractions.
Dissertation (MEng)--University of Pretoria, 2008.
Electrical, Electronic and Computer Engineering
unrestricted