Search

Ski jumping, a prestigious Olympic discipline, has been constantly evolving. Over the years, the biomechanics fundamentals of this highly technical sport have been intensively documented by investigating the jump timing, kinematics, dynamics and motor control. This has increased our understanding of this discipline and resulted in the enhancement of the performance, in the development of new jumping techniques and in implications in safety and fairness. However, the performed analyses, based on standard measurement devices (e.g., 3D camera-based system, force plate), suffered from their complexity, time-consumption and related expenses. Consequently, they have been restricted to few athletes and focused on specific portions of the jump. On the other hand, to be compliant with the coaches’ and athletes’ needs, performance evaluation in daily routine relies primarily on feeling, visual observation and standard video analysis, but this type of evaluation suffers from subjectivity. Thus, nowadays, both scientists and field experts agree that there is a lack of an objective and simple assessment method for ski jumping. In this thesis, we propose and assess a new wearable inertial-based system and dedicated methods usable in training practice to extract timing, angular, dynamic, and coordination parameters during the entire jump. Then, we use the system to characterize the entire ski jump and evaluate the key parameters relative to performance using a relatively large cohort of athletes. The proposed measurement tool was devised to be easy-to-handle by coaches or athletes without the need for technical assistance. Within five minutes, the system was set up on the athlete and didn’t require any calibration constraints. The jumper was then ready for training as usual. After the training session, data were downloaded and then processed automatically in few seconds with the use of appropriate software. The key temporal events and related ski jumping phases were automatically detected on the raw inertial signals using a pattern recognition technique. A new algorithm was designed to monitor robust and drift-free orientations of skis and body segments for the entire jump sequence using a fusion of the inertial signals and specific biomechanical constraints during the jump. The dynamics of the take-off was evaluated using a forward kinematics model, while the flight aerodynamics was directly assessed using the inertial signals. The continuous relative phase was proposed to characterize the inter-segment coordination of lower-limbs during take-off. To assess the jump performance, traditional as well as new parameters describing the timing, the kinematics, dynamics and coordination were extracted. The proposed methods were assessed by their validity against reference measures when possible, by their consistency with the literature, and by their sensitivity to discriminate performance changes using athletes of different levels. To characterize the entire ski jumping sequence, 33 athletes, ranging from junior to world-class, were monitored during training conditions (in total, 87 jumps recorded). The temporal, angular, coordination, and dynamic classes of parameters were extracted using the wearable inertial system for the take-off, early-flight, and stable flight, and they were compared to performance (the jump length) using different statistical approaches, i.e. bivariate correlation, multiple regression (partial least-square regression), and factor analysis. It showed that ski jumping is a complex sequence of movements involving numerous parameters. This thesis proposed and assessed original inertial-based methods to evaluate the timing, kinematics, dynamics, and coordination in ski jumping. The measurement tool automatically extracts valid and sensitive parameters to characterize the performance and provides new insights by analysing the movement over the entire jump sequence. Giving a quantitative, rapid and personalized feedback with this new simple wearable system will help the athletes and coaches to better assess specific skills and improve the training efficiency.

Parkinson's disease (PD) is the second most common neurodegenerative disease in the general population. Cardinal symptoms of Parkinson's disease are resting tremor, rigidity, akinesia and bradykinesia and in advanced stages, gait impairments, postural instability and complications of chronic treatment with levodopa such as motor dysfunctions and dyskinesia. Multitude and complexity of these motor symptoms and their variability over the time have made assessment of them a difficult task. Moreover, following the fluctuations of motor performance (ON/OFF fluctuations) of the PD patients throughout their daily activities by quantifying their motor symptoms is a major challenge. The aim of this thesis was to design and validate a portable ambulatory movement analysis system for long-term monitoring and qualitative and quantitative assessment of motor abnormalities of PD patients during daily activities. We have designed a new measurement system consisting of five independent, lightweight, autonomous sensing units based on kinematic sensors that can continuously record body movements during daily life. Using this system and by performing several clinical studies, both in controlled conditions and on free moving patients, we have prepared a database of different movement patterns of PD patients. This database was the basis to design several new algorithms for the analysis of tremor, bradykinesia, gait and posture. An accurate algorithm based on spectral estimation has been proposed to detect and quantify tremor during daily activities of PD patients with a resolution down to three seconds using gyroscopes attached to the forearms. By quantifying the speed, range and the frequency of the movements, we have proposed a new method to assess the bradykinesia and tested it both in controlled and free conditions. We found out that in the free moving patients, the outcomes of this algorithm show significant and good correlation to the established clinical scores. Regarding the detection and analysis of gait, we have developed and tested a method based on four sensors attached to the lower limbs that provided spatio-temporal parameters of gait with good accuracy. We further improved our method using a new biomechanical model that could predict the movements of thighs from the movements of shanks during walking. This way we could reduce the number of sensor sites on the body while keeping the same accuracy in estimation of the spatio-temporal parameters of gait. By combining a statistical classifier, to detect transitions between sitting and standing postures, and a fuzzy classifier, to detect the basic body postures, we have developed an algorithm to classify basic body posture allocations both in PD patients and aged matched healthy subjects. Finally, while currently no other objective ambulatory method exists to accurately detect the periods of ON and OFF in PD patients, by combining the outcomes of the above algorithms (tremor, gait, bradykinesia and posture) using a statistical approach, we have proposed a method to detect periods of these two states with a resolution of 10 minutes in free moving patients. We believe that the proposed system has a high potential both for the clinical applications and research purposes related to the patient with Parkinson's disease and possibly other neurological movement disorders.

Stroke, or cerebral-vascular accident is the leading cause of disabilities in the western world. It primarily affects mobility due to a degradation of principal motor functions such as balance and gait. This reduction of mobility directly impacts the spectrum of possible activities of daily-living and consequently the level of independence. Patients undergo intensive physical therapy for improving various aspects of mobility. To evaluate the rehabilitation strategy outcome and guide patients through the therapy, clinical instruments have been widely used including questionnaires and motor function tests. Questionnaires related to mobility and quality of life depend on the interpretation, the cognitive abilities and the state-of-mind of the patient during the interview and are therefore subject to the interpretation of the patient. Motor function tests, such as Berg Balance Scale and Timed-Up-and-Go for evaluating balance and gait disorders, evaluates the patient’s motor capacity of performing given motor tasks in clinical settings, under close supervision. They are consequently not necessarily representative of the patient’s motor performance in daily life. Providing a measure of mobility performance would allow clinicians to better understand the impact, in patient’s home environment, of various rehabilitation strategies. Furthermore, emerging rehabilitation techniques such as virtual reality games may evaluate the patients during the training to build a progression scheme. However, the in-game assessment is limited to their scope of rehabilitation, i.e. it is restricted to one single component of mobility. To overcome the limitations of current assessment tools, an objective evaluation tool of stroke patient’s recovery in their daily-life environment is required. Recent miniaturization of wearable technologies have enabled new possibilities for long-term monitoring of physical activity in daily life. The detailed measurement of physical activity includes: the characterization of activity events such as postural (sit-to-stand and stand-to-sit) transitions; the quantification of the intensity and the amount of activities; and a pattern analysis of these events/activities through their distributions and temporal structures. The aim of this thesis is to provide an objective assessment tool of stroke patients’ mobility performance in daily-life. According to the work performed, the manuscript is organized in four main parts: (i) selection of an appropriate barometric pressure sensor for improving postural transitions recognition, with validation on a group of healthy volunteers; (ii) improvement of a postural transition recognition in stroke patients using the selected barometric pressure sensor; (iii) design of an activity classification algorithm suitable for stroke patients by fusing barometric pressure and inertial sensor information; (iv) cross-sectional comparison (elderly controls vs stroke patients) of the mobility metrics (transition, activity, and pattern metrics) and their relationships with motor function tests. The wearable system devised in this thesis provides clinicians with a convenient (one sensor on the trunk) and more complete monitoring solution able to characterize the various aspects of stroke’s mobility in daily-life. This solution can be used to objectively and accurately assess the rehabilitation effectiveness and eventually to adapt/fit the rehabilitation program to the patient's needs.

As one of the three most popular sports in the Summer Olympics, competitive swimming has always been an attractive subject of study for sports scientists. The intricate nature of the swimmer's movements and the variety of techniques have led coaches to require analysis systems to gain a more detailed understanding of swimmers' performances. Surveys have shown that despite recent technological advances in sports, swimming coaches still need a measurement device that is accessible, easy to use, and provides easy-to-understand results. Conventional analysis systems such as high-resolution cameras, while accurate, are too time-consuming and cumbersome for daily use. With the advent of wearable sensors, especially inertial measurement units (IMU), motion analysis has gained the ability to not only study new aspects of motion, but also cover in-field applications. However, researchers focused primarily on extracting features rather than using them to evaluate performance and provide feedback in the field. Free-swimming is the phase that has been most studied with IMUs, and among the major swimming styles, front crawl has attracted more attention. Therefore, a comprehensive analysis approach based on IMUs covering all swimming styles and phases can provide the swimming community with deeper insight into swimmers' performance and a better solution to training needs. This thesis presents a novel methodology for swimming analysis based on inertial wearable sensors. The proposed method uses a unified macro-micro analysis approach that scans the entire training session for swimming bouts and then narrows down to separating the swimming phases from wall to wall. The method is implemented using the IMU data from different sensor positions on the swimmer's body for comparison. Based on the results of the developed algorithms, sacrum position was determined to be optimal for detecting all swimming phases. As a result, the analysis then detects a set of spatio-temporal and kinematic parameters based on the IMU data on the sacrum at each phase, which are used for the swimmers' phase-based performance evaluation. Furthermore, the extracted parameters were shown to reflect different aspects of the swimmer's performance, such as propulsion, posture, or efficiency. The developed performance evaluation method estimates a set of velocity-based goal metrics, validated against the reference system, that represent how well the swimmer performed during the corresponding phase. Finally, the system is used to provide weekly feedback to a team of young swimmers to evaluate the application of feedback in practice and its impact on swimmers' progress. Overall, this thesis proposes a new approach to swimming performance evaluation based on IMUs, with a broader scope of application to all phases and styles of swimming. It aims to expand the application of IMUs in swimming by providing spatio-temporal and kinematic parameters that represent critical aspects of swimming and objectively evaluate swimmers' performance. The estimated goal metrics are sensitive to swimmers' progress during weeks of training. In addition, the potential of such an analysis system for in-field training sessions is tested and showed promising results. Finally, coaches can obtain a more detailed view of swimmers' techniques at both macro and micro levels using a single IMU sensor on a daily basis.

In orthopaedics, outcome evaluation is currently recognized as a major challenge to improve the knowledge both about injuries and treatments. For anterior cruciate ligament (ACL) rupture, various evaluation tools are available for routine uses: clinical examinations, static knee joint laxity measurements and scores. Although these tools can diagnose an ACL rupture and evaluate the results of a surgery, they are inadequate to answer the current therapeutic issues. Therefore, since two decades, gait analysis was investigated as an objective evaluation approach to quantify more subtle kinematics modifications. However, these analyses need dedicated laboratories including complex and expensive devices for motion capture. Consequently, only few studies focusing on ACL were conducted up to now. These studies enrolled a limited number of patients and reported partly divergent results. Thus, nowadays, surgeons lack a reliable, objective and simple method to evaluate the three-dimensional (3D) knee joint function in field during daily activities and conditions. In this thesis, we propose and assess new devices and algorithms to fulfil these needs. The contributions are driven by four main objectives: to provide appropriate measurement systems, to ensure repeatable quantification according to clinical descriptions, to determine the suitability of the kinematic parameters for outcome evaluation, and to use these new tools in clinical applications, both to show their relevance and provide new insights about the lower limb 3D function. Measurement systems. Three devices are designed in this thesis. The first device consists of a portable data logger and two 3D gyroscopes fixed on the thigh and the shank segments. It quantifies the 3D function of the knee joint. The second system fuses 3D gyroscopes with 3D accelerometers in order to measure the orientations during short periods. By strapping several of these sensors units on the lower limb segments, this system allows ambulatory measurement of 3D kinematics. Finally, an original system combining a portable magnetic tracker and inertial sensors is proposed to measure body segments orientation over long durations. The performances of all systems are assessed against high level reference devices. Functional calibrations. In order to quantify the 3D function of the knee joint independently of the sensors location on the segments, calibration methods suitable for the inertial-based systems are proposed. These methods follow the recommendations of the International Society of Biomechanics (ISB) and avoid the localization of anatomical landmarks. They rely on predefined movements and posture measurements. The angular displacements obtained through these functional calibration procedures when collecting the data with one of the proposed system are positively comparable to previous systems in the literature. Kinematic parameters characterisation. Before using any data to quantify kinematics changes induced by an injury or a treatment, it is essential to evaluate their reliability. The propagation of both, anatomical-based and functional calibration procedures errors to 3D kinematics is investigated. The repeatability of the parameters is also evaluated. Finally, the effects of soft tissues artefacts are estimated using an exoskeleton harness. Clinical applications. The evaluation methods proposed here are involved in three different clinical protocols, both to evaluate their convenience of use and to contribute to the increase of clinical knowledge. Two applications enrol ACL-deficient patients prior and following reconstructive surgery, and a third one considers four groups of patients with various ankle conditions. During these studies, long distance ground level, and incline walking activities are monitored. This thesis proposes original inertial-based methods to evaluate the 3D function of the knee joint. The measurement errors are comparable to those of standard in lab devices, while the new systems allow ambulatory monitoring in clinical field. The characterization of the kinematic parameters confirms the value of the literature when existing. Finally, the clinical applications report promising results and validate the suitability of the methods for routine uses.

Locomotion is one of the most important abilities of humans. Actually, gait locomotion provides mobility, and symbolizes freedom and independence. However, gait can be affected by several pathologies, due to aging, neurodegenerative disease, or trauma. The evaluation and treatment of mobility diseases thus requires clinical gait assessment, which is commonly done by using either qualitative analysis based on subjective observations and questionnaires, or expensive analysis established in complex motion laboratories settings. This thesis presents a new wearable system and algorithmic methods for gait assessment in natural conditions, addressing the limitations of existing methods. The proposed system provides quantitative assessment of gait performance through simple and precise outcome measures. The system includes wireless inertial sensors worn on the foot, that record data unobtrusively over long periods of time without interfering with subject's walking. Signal processing algorithms are presented for the automatic calibration and online virtual alignment of sensor signals, the detection of temporal parameters and gait phases, and the estimation of 3D foot kinematics during gait based on fusion methods and biomechanical assumptions. The resulting 3D foot trajectory during one gait cycle is defined as Foot Signature, by analogy with hand-written signature. Spatio-temporal parameters of interest in clinical assessment are derived from foot signature, including commonly parameters, such as stride velocity and gait cycle time, as well as original parameters describing inner-stance phases of gait, foot clearance, and turning. Algorithms based on expert and machine learning methods have been also adapted and implemented in real-time to provide input features to recognize locomotion activities including level walking, stairs, and ramp locomotion. Technical validation of the presented methods against gold standard systems was carried out using experimental protocols on subjects with normal and abnormal gait. Temporal aspects and quantitative estimation of foot-flat were evaluated against pressure insoles in subjects with ankle treatments during long-term gait. Furthermore, spatial parameters and foot clearance were compared in young and elderly persons to data obtained from an optical motion capture system during forward gait trials at various speeds. Finally, turning was evaluated in children with cerebral palsy and people with Parkinson's disease against optical motion capture data captured during timed up and go and figure-of-8 tests. Overall, the results demonstrated that the presently proposed system and methods were precise and accurate, and showed agreement with reference systems as well as with clinical evaluations of subjects' mobility disease using classical scores. Currently, no other methods based on wearable sensors have been validated with such precision to measure foot signature and subsequent parameters during unconstrained walking. Finally, we have used the proposed system in a large-scale clinical application involving more than 1800 subjects from age 7 to 77. This analysis provides reference data of common and original gait parameters, as well as their relationship with walking speed, and allows comparisons between different groups of subjects with normal and abnormal gait. Since the presented methods can be used with any foot-worn inertial sensors, or even combined with other systems, we believe our work to open the door to objective and quantitative routine gait evaluations in clinical settings for supporting diagnosis. Furthermore, the present studies have high potential for further research related to rehabilitation based on real-time devices, the investigation of new parameters' significance and their association with various mobility diseases, as well as for the evaluation of clinical interventions.

The importance of outcome evaluation of a medical treatment in orthopedics is currently recognized. In shoulder disease, a large variety of evaluation tools is employed to assess the results of the surgery. However, even if the majority of these evaluations are largely widespread, none was accepted as a universal standard. Since 1990, few researchers have been evaluating the assumption that the movement analysis (with camera-based or electromagnetic systems) is likely to provide objective results. In clinical practice, these techniques are not always applicable for outcome evaluation of a treatment. The surgeons lack a convenient and simple method of evaluating in an objective way a patient's activity and quality of life after a surgery of the shoulder. This project provides a new tool for the objective functional evaluation of shoulder pathologies, a tool that can be easily used by a doctor at a hospital and by the patient at home. It allows the measurement of the biodynamic changes as well as 3D kinematics of the treated shoulder by noting the effects of these changes on clinical results and on the patient's daily activity. The project was split in four complementary studies. In the first study, a new ambulatory device allowing long-term monitoring of the shoulder movement using several inertial sensors (3D gyroscopes, 3D accelerometers) attached on the trunk, the humerus and the scapula's spine was designed. By combining acceleration and angular velocity features of the both humerus during 9 tests, three kinematic scores for the functional assessment of the shoulder were presented to evaluate the shoulder function in patient before and after surgery. The kinematic scores objectively showed the shoulder improvement after surgery. In the second study, a new method was proposed to detect and quantify the dominant upper-limb segment during daily activity. The method was tested on healthy subjects (N=31) and a patient group (N=10, at baseline, 3, 6 and 12 months after surgery) while carrying the system during 8 hours of their daily life. The results showed the dominance of the arm during standing, sitting and walking periods for healthy subjects and the quantification of the shoulder improvement after surgery, by taking into account the presence of the disease in the dominant or the non dominant arm. In the third study, 3D gyroscopes attached on the humerus were used to identify the movements of flexion-extension, abduction-adduction and internal/external rotation of the humerus and to identify the rates of adjunct (deliberate rotation) and conjunct rotations (inherent or automatic rotation) within each movement. The frequencies of each movement (number/hour) for the different ranges of the arm speed, as well as the rate of adjunct and conjunct rotations for each movement were estimated during daily activity in healthy and patient groups. The results provided the values of frequency of each movement and adjunct/conjunct rate based on the data obtained from the healthy group. In the pathological case, we found that the painful dominant shoulder of the patients lost its predominance in favor of the healthy shoulder, the non dominant shoulder. Patients had less pure internal/external rotations and performed less fast movements while after surgery these parameters presented no significant differences with the healthy group. In the fourth study, a new method of detecting the working level of the shoulder was presented. By measuring the arm elevation during motionless periods, we proposed a new score to evaluate the ability of working at a specific level for a definite duration. We showed that this score had an average of 100% (±31%) for healthy subjects while the working level of the painful shoulder was lower than the healthy shoulder and improved significantly after surgery (up to 87% at 6 months). This study provides preliminary evidence of the effectiveness of the proposed system in clinical practice and objectively assesses upper-limb activity during daily activity.

Balance impairment is one of the major contributors to fall, frequently amenable to surgical intervention. Repeated falls can lead to bad injuries which can decrease or even stop the professional and social activities. The study of the static equilibrium is an easy method to quantify the physical/functional status of a person, which can help physicians to establish the diagnostic. In this study, body sway was assessed using 3D inertial sensors (accelerometers and gyroscopes). Based on recorded kinematical signals under different measurement protocols (e.g. eyes open/eyes closed), the goal was to find parameters able to objectively discriminate subjects with low equilibrium skills from the others. The kinematical equilibrium parameters were estimated using different linear/nonlinear signal processing methods. The results of eyes open/eyes closed measurements gave a better understanding of the interaction of the biological sensors used by the body to regulate the equilibrium. The linear analysis study provided a first approach of the main kinematical parameters to discriminate the participants in the study. Finally, the nonlinear analysis provided new parameters discriminating better the physical performances of participants even if some overlap in the participants' data persist

Following the cultural revolution of the late 1960s, the number of elite and recreational runners rose consistently, reaching approximately 7.9 million road races participants in 2018. Today, running is everywhere. City parks, forests, mountain trails, and athletic tracks are now the playground of numerous running enthusiasts, whatever their ages, gender, and social background. With such heterogeneity in the runners' profiles, the motives to maintain a running habit vary from psychological, social, and physical objectives to performance-oriented goals. Although the health benefits of running are well-recognized in the scientific literature, its regular practice also presents risks of injuries. To study the underlying mechanisms associated with injuries or improvements in performances, scientists have investigated the kinematics and the kinetics characteristics of the running gait. Habitually, this quest requires the use of precise monitoring instruments only accessible in well-equipped research laboratories. However, over the past two decades, the advent of wearable sensors shifted the analysis of running into real-world settings, where runners encounter different environments, outside, in the wild. It is in this setting that the current thesis situates itself. This thesis presents a new wearable system for the objective assessment of the running gait in real-world conditions. The proposed method uses foot-worn inertial sensors and lab-validated algorithms to provide a reliable analysis of the spatiotemporal parameters of running. The system can operate outdoors and over extended periods while providing a quasi-real-time evaluation of each step. Further, with its automatic detection of the sensor location and calibration, the proposed method is easy-to-use and accessible to non-initiated users. For the technical validation of the proposed system, the spatiotemporal metrics were compared with gold-standard reference systems. Temporal events and gait phases were validated in-lab against a reference force plate integrated into a treadmill, and the results of a novel orientation-drift correction model compared to a state-of-the-art motion capture system. Three overground speed estimation methods were evaluated in real-world conditions and compared to a Global Navigation Satellite System device. Finally, these methods were tested in different settings, such as a marathon race, a mountain ultra-marathon, and a 400-m hurdling competition. These tests provided valuable insight into the limitations of the proposed system and suggested several improvements for its use in real-world conditions. Overall, this thesis aims to augment the resolution of running analysis by handling the technical challenges associated with inertial sensors and providing fast and reliable biomechanical metrics. As such, the system could contribute to extending the knowledge about the mechanical adaptations experienced in real-world environments and long-term running. Moreover, the potential of such an instrument for in-field performance evaluation was tested in this thesis and showed promising results. Hence, such an assessment of the running gait during training and competitions could help athletes and coaches monitor the training load and improvement in performances. Finally, with the advents in the miniaturization of wearable sensors, the proposed methods could be used in various running-related applications, such as shoe-fitting or the analysis of other sports.

Muscle weakness is a common symptom in elderly population and in children with Duchenne muscular dystrophy (DMD), causing impaired mobility. DMD is a devastating degenerative disease caused by a mutation in the dystrophin gene, leading to the absence of the corresponding protein. Children with DMD experience with disease progression a gradual decline in their ability to perform functional tasks. Precise methods of measurement are necessary to more accurately establish the effect of treatment and find the temporal trend in functional motor status. With regard to the aging population, it is largely recognized that there is a substantial need for an accurate health assessment. Thus, it can have significant implications for older patients' care planning and future quality of life. Despite ongoing research, the development of reliable and objective outcome measures in this population is still a major challenge. The development of these outcome measures is further complicated by certain aspects such as frailty, comorbidity, and heterogeneity in this population. The assessment of health's status in aged population and DMD children is not only useful for the prediction of health decline with age, but also for the evaluation of the outcomes of early treatments in both populations. In DMD, large research efforts are made to find efficient treatments to slow/stop the disease progression. Finding appropriate outcome measures which reflect an improvement or stabilization in functional ability of the subject is also another challenge that clinicians are currently facing. The research aims of this thesis were: (i) to provide new and accurate measures/outcomes in order to assess the effect of treatment and to determine the temporal trend in functional motor status in two types of disorders related to muscle weakness (frailty and Duchenne muscular dystrophy); (ii) to analyze relevant body movements such as postural transitions and gait in laboratory as well as in real-world conditions/environment; (iii) to compare the objective outcomes obtained by analyzing body movements to the relevant clinical scores. New objective parameters have been proposed to quantify the sit-to-stand and stand-to-sit movements. In regard to the one of the most functional demanding daily tasks which is the ability to rise from a chair (sit-to-stand transition), these parameters showed high sensitivity to change. From a clinical point of view, they separated successfully frail elderly subjects from healthy elderly subjects, as well as different evaluations of frail elderly before and after the rehabilitation program. The new parameters developed in the current study may capture more subtle functional changes that might help for prediction of adverse events during the disability development. A new algorithm was designed for detection of the sit-to-stand and stand-to-sit postural transitions patterns in real-life conditions based only on one body-worn inertial sensor. For the first time a technical validation was proposed with real-world data and its superiority compared to existing controlled validation protocols was shown. The proposed algorithm provides a simple method to detect and characterize postural transitions in healthy subjects and also in disorders such as chronic pain and frail elderly subjects. We showed that gait parameters obtained during long distance walking separated successfully the DMD patients from age-matched healthy children. The study showed the potential of such parameters to distinguish between different stages of the disease and open new perspectives for an objective assessment of the efficacy of some new therapies associated with Duchenne muscular dystrophy. We believe that these research investigations may have applicability in both clinical and research fields related to elderly, DMD children and probably other neuromuscular disorders.

The importance of evaluation of an orthopedic operation such as hip or knee arthroplasty has long been recognized. Many definitions of outcome and scoring questionnaires have been used in the past to assess the outcome of joint replacement. However, these assessments are subjective and not accurate enough. In addition, orthopedic surgeons require now more subtle comparisons between potentially efficacious treatments (e.g. two types of prostheses). Therefore, the use of objective instruments that have a better sensitivity and specificity than traditional scoring systems is needed. Gait analysis is one of the most currently used instrumented techniques in this respect. However, a gait analysis system is accessible only in a few specialized laboratories, as it is complex, expensive, need a lot of room space and fixed devices, and not convenient for the patient. In this thesis, we proposed an ambulatory system based on kinematic sensors attached on the lower limbs to overcome the limitations of the previously mentioned techniques. Technically the device is portable, easily mountable, non-invasive, and capable of continuously recording data in long term without hindrance to natural gait. The goal was to provide gait parameters as a new objective method to assess Total Knee Replacement (TKR). New solutions to fusing the data of accelerometers and gyroscopes were proposed to accurately measure lower limbs orientations and joint angles. The methods propose a minimal sensor configuration with one sensor module mounted on each segment. The models consider anatomical aspects and biomechanical constraints. In the proposed techniques, the angles are found without the need for integration, so absolute angles can be obtained which are free from any source of drift. These data were then used to develop a gait analysis system providing spatio-temporal parameters, kinematic curves, and a visualization tool to animate the motion data as synthetic skeletons performing the same actions as the subjects. Moreover, a new algorithm was proposed for assessing and quantification of inter-joint coordination during gait. The coordination model captures the whole dynamics of the lower limbs movements and shows the kinematic synergies at various walking speeds. The model imposes a relationship among lower limb joint angles (hips and knees) to parameterize the dynamics of locomotion for each individual. It provides a coordination score at various walking speeds which is ranged between 0 and 10. An integration of different analysis tools such as Harmonic Analysis, Principal Component Analysis, and Artificial Neural Network helped overcome high-dimensionality, temporal dependence, and non-linear relationships of the gait patterns. In order to show the effectiveness of the proposed methods in outcome evaluation, we have considered a clinical study where the outcomes of two types of knee prostheses were compared. We conducted a randomized controlled study, including 54 patients, to assess TKR outcome between patients with fixed bearing and mobile bearing tibial plates of implants. The patients were tested preoperatively and postoperatively at 6 weeks, 3 months, 6 months, and 1 year. Various statistical analyses were done to compare the outcomes of the two groups. Finally, we provided objective criteria, using ambulatory gait analysis, for assessing functional recovery following TKR procedure. We showed significant difference between the two groups where the standard clinical evaluation was unable to detect such a difference.

Two surgical treatments exist for ankle osteoarthritis (AO): ankle arthrodesis (AA) and total ankle replacement (TAR). Due to advantages and drawbacks of each treatment, decision making about these treatments is still a clinical question and thus, their outcome evaluation is of clinical interest. So far, outcome evaluation has been done via clinical scales based on questionnaires and clinical observations. However, clinical scales are subjective and are not accurate enough to detect subtle alterations in foot function. Therefore, objective outcome tools such as instrumented gait analysis are advantageous for evaluation of ankle osteoarthritis and its surgical treatments. In particular, possible alterations of kinematics and kinetics of ankle and other foot joints during gait can be relevant for outcome evaluation of these treatments. On the other hand, gait analysis has been performed in gait laboratories using complex and expensive equipments which are not available in all clinics. Besides, limited measurement space of gait laboratories can perturb the natural gait. Therefore, there is a need for in-field gait analysis for outcome evaluation of ankle osteoarthritis treatments. The aim of this thesis is to design and validate new ambulatory systems for kinematics and kinetics assessment of multi-segment foot during long-distance gait and to apply these systems for objective outcome evaluation of ankle osteoarthritis treatments. The present study proposed objective kinematic criteria for foot segmentation. According to these criteria, for clinical evaluations, foot and ankle complex could be reliably divided into shank, hindfoot, medial forefoot, lateral forefoot and toes. Then, in order to choose an appropriate joint kinematics descriptor, the common mathematical tools to describe the kinematics of foot joints were evaluated and showed similarly high consistency in their results and enough low sensitivity to measurements errors. New algorithms were introduced to estimate rotation of foot joints, 3D ground reaction force and kinetics of foot joints (force, moment and power) using ambulatory systems. First, kinematics of multi-segment foot was assessed, based on body-fixed inertial sensors (3D gyroscope ad 3D accelerometer) on shank, hindfoot, forefoot and toes. Second, plantar pressure insole, as an available ambulatory system in clinics, was used for assessing 3D ground reaction force. An optimal algorithm was proposed for estimating the 3D ground reaction force based on plantar pressure distribution. Third, inertial sensors and pressure insole were used for ambulatory assessment of multi-segment foot joints kinetics. These designed systems were validated with gold standard reference systems. As an advantage, they can be easily used in different clinical environments for long-distance field measurement without perturbing the natural gait. After technical validation of the designed measurement systems, their suitability and efficiency for clinical evaluations were investigated within a clinical protocol. Four populations were involved in the study: patients with AO, patients with TAR, patients with AA and healthy subjects. Gait parameters were measured by ambulatory systems in long-distance walking and were compared between patient groups and healthy subjects. In general, kinetics of foot joints and pressure parameters showed alteration of foot function after AO and were not completely restored after both treatments. Rotation of joints showed improvement only after TAR. Spatio-temporal parameters of gait and their variability showed both of the mentioned effects as well as improvements after both TAR and AA. Therefore, the proposed parameters could detect the differences between patients and healthy subjects before and after both treatments. These parameters also showed clinical relevance based on their correlation with clinical scales. In this thesis, using an ambulatory system consisted of inertial sensors and plantar pressure insoles, a new way of gait analysis to evaluate the function of the foot and ankle complex was proposed and validated. The results provided pertinent metrics for objective outcome evaluation of ankle osteoarthritis treatments and confirmed the suitability of this system for clinical routine uses.

Daily activity reflects the health status of an individual. Ageing and disease drastically affect all dimensions of mobility, from the number of active bouts to their duration and intensity. Performing less activity leads to muscle deterioration and further weakness that could lead to increased fall risk. Gait performance is also affected by ageing and could be detrimental for daily mobility. Therefore, activity monitoring in older adults and at risk persons is crucial to obtain relevant quantitative information about daily life performance. Activity evaluation has mainly been established through questionnaires or daily logs. These methods are simple but not sufficiently accurate and are prone to errors. With the advent of microelectromechanical systems (MEMS), the availability of wearable sensors has shifted activity analysis towards ambulatory monitoring. In particular, inertial measurement units consisting of accelerometers and gyroscopes have shown to be extremely relevant for characterizing human movement. However, monitoring daily activity requires comfortable and easy to use systems that are strategically placed on the body or integrated in clothing to avoid movement hindrance. Several research based systems have employed multiple sensors placed at different locations, capable of recognizing activity types with high accuracy, but not comfortable for daily use. Single sensor systems have also been used but revealed inaccuracies in activity recognition. To this end, we propose an instrumented shoe system consisting of an inertial measurement unit and a pressure sensing insole with all the sensors placed at the shoe/foot level. By measuring the foot movement and loading, the recognition of locomotion and load bearing activities would be appropriate for activity classification. Furthermore, inertial measurement units placed on the foot can perform detailed gait analysis, providing the possibility of characterizing locomotion. The system and dedicated activity classification algorithms were first designed, tested and validated during the first part of the thesis. Their application to clinical rehabilitation of at risk persons was demonstrated over the second part. In the first part of the thesis, the designed instrumented shoes system was tested in standardized conditions with healthy elderly subjects performing a sequence of structured activities. An algorithm based on movement biomechanics was built to identify each activity, namely sitting, standing, level walking, stairs, ramps, and elevators. The rich array of sensors present in the system included a 3D accelerometer, 3D gyroscope, 8 force sensors, and a barometer allowing the algorithm to reach a high accuracy in classifying different activity types. The tuning parameters of the algorithm were shown to be robust to small changes, demonstrating the suitability of the algorithm to activity classification in older adults. Next, the system was tested in daily life conditions on the same elderly participants. Using a wearable reference system, the concurrent validity of the instrumented shoes in classifying daily activity was shown. Additionally, daily gait metrics were obtained and compared to the literature. Further insight into the relationship between some gait parameters as well as a global activity metric, the activity “complexity”, was discussed. Participants positively rated their comfort while using the system... (Please refer to thesis for full abstract)

Nowadays inertial sensors are extensively used for gait analysis. They can be used to perform temporal event detection (i.e. step detection) and to estimate the orientation of the feet and other body segments to determine walking speed and distance. Usually, orientation is estimated from integration of the measured angular velocity. Prior to integration of measured acceleration to obtain speed, the gravity component has to be estimated and removed. During each integration small measurement errors accumulate and result in so-called drift. Since the first uses of inertial sensors for gait analysis methods have been presented to model, estimate and remove the drift. The proposed methods worked well for relatively slow movements and movements taking place in the sagittal plane. Many methods also relied on periodically occurring static phases such as the stance phase during walking to correct the drift. Inertial sensors could also be used to track higher dynamic movements, for example in sports. Potential applications focus on two aspects: performance analysis and injury prevention. To better explain and predict performance, in-field measurements to assess the coordination, kinematics, and dynamics are key. While traditional movement analysis (e.g. video analysis) can answer most of the questions related to both performance and injury, they are cumbersome and complex to use in-field. Inertial sensors, however, are perfectly suited since they allow to measure the movement in any environment and are not restricted to certain capture volumes. Nevertheless, most sports have very high movement dynamics (e.g. fast direction changes, high speeds) and are therefore challenging for computing reliable estimates of orientation, speed and position. The inertial measurements are compromised by noise and movements oftentimes don't provide static or slow phases used in gait analysis for drift correction. Therefore, the present thesis aimed to propose and validate new methods to model, estimate and remove drift in sports and for movements taking place outdoors in uncontrolled environments. Three different strategies were proposed to measure the movement of classical cross-country skiing and ski mountaineering, alpine ski racing, and outdoor walking over several kilometres. For each activity specific biomechanical constraints and movement dynamics were exploited. The proposed methods rely only on inertial sensors and magnetometers and are able to provide orientation, speed, and position information with an accuracy and precision close to existing gold standards. The most complete system was designed in alpine ski racing, probably one of the most challenging sports for movement analysis. Extreme vibrations, high speeds of over 120 km/h and a timing resolution below 0.01 seconds require maximum accuracy and precision. The athlete's posture and the kinematics of his centre of mass both in a relative athlete-centred frame and in a global Earth-fixed frame could be obtained with high accuracy and precision. Where 3D video analysis requires a very complex experimental setup and takes several hours of post processing to analyse a single turn of a skier, the proposed system allows to measure multiple athletes and complete runs within minutes. Thus, new experimental designs to assess performance and injury risk in alpine ski racing became feasible, greatly helping to gain further knowledge about this highly complex and risky sport.

Quantification of mobility is the key to monitor the progression of mobility disorders as well as the effect of an intervention. Inertial measurement units (IMUs) with dedicated algorithms can quantify postural transitions and gait as the two key aspects of mobility in an objective and continuous manner. IMU-based mobility assessments can be performed by either functional tests in the clinic or through daily activities. Assessments performed in the clinic are more indicative of peoples best performance or capacity, while assessments performed at home represent mostly their actual performance. Yet the relationship between these two settings is not fully understood, both due to the existing gaps in technical algorithms as well as challenges in comparing two inherently different domains. To this end, in this thesis, I firstly focused on developing and validating algorithms to quantify mobility in both clinical and domestic environments. The added clinical value of these IMU-based mobility assessments was shown in several populations with mobility impairments. Finally, by proposing novel approaches, I focused to bridge the gap between clinical and daily activity assessments. The previous approaches to quantify mobility are mostly based on algorithms that are validated only during clinical or lab-based assessments. Opposed to daily activities, lab assessments contain simple and single-task activities. Therefore, it is important to design algorithms robust to the complex context of daily life setting while being unobtrusive to daily activities. A new algorithm was introduced to detect and characterize postural transitions, i.e., sitting and standing. Next, machine-learning-based algorithms were developed to detect walking bouts and estimate gait speed. The proposed postural transition and gait quantification algorithms were based on a single IMU on the lower back which is unobtrusive to daily activities. The novelty of the algorithms is their robustness to sensor placement changes during daily activities. The proposed algorithms demonstrated high performance during both clinical and daily activity assessments whether on healthy individuals or participants with mobility impairments. Next, through several analyses, I demonstrated how such instrumented mobility assessments can discriminate different patient populations. For instance, IMU-derived mobility parameters could differentiate older adults with and without risk of falls as well as patients with moderate or severe stages of multiple sclerosis. Moreover, the aforementioned parameters were compared between clinical and daily activity assessments. By this comparison, clinicians can have a better understanding of patients capacity through a remote assessment of mobility. Finally, it was shown how clinical and daily activity assessments can provide complementary information to each other. For instance, by introducing novel approaches to compare gait speed between clinical and daily activity assessments, the effect of the medication in Parkinsons disease (PD) was traceable during daily activities. The findings can lead to better optimization of the medication dose in PD. Overall, this thesis provided a framework that can help clinicians with an objective assessment of mobility. Furthermore, the approaches introduced in this thesis can help for better management of intervention and tracking its effects where both clinical and daily activity assessments exist.

The shoulder is the joint with the largest range of motion in the human body. It is composed of three articulations, the glenohumeral articulation providing about two thirds of its mobility which is, also the most complex and fragile of the three. For this reason, it possesses a particularly complex stabilization mechanism involving a large number of different structures, making it particularly difficult to model, evaluate and treat. Since many structures are involved in the stabilization of this articulation, pathologies affecting any of these elements may compromise the whole stabilization mechanism. In that case, a good understanding of the mechanisms involved, combined with reliable evaluation tools are necessary to provide an effective treatment. At present, different modelling approaches tried to model the phenomenon of glenohumeral instability, but none was able to replicate the whole mechanism reliably. These limitations are particularly problematic during total shoulder replacement surgeries, where the inability to reproduce healthy glenohumeral motions is the main cause of implant failure. The development of a model able to predict the onset of GH instability based on the surgical planning, would greatly improve the general survivor-ship of the implants, and is one of the main goals of this thesis. Additionally, the clinical tools available for the evaluation of the functional outcome are limited. Clinical questionnaires, static apprehension and laxity scores and two-dimensional evaluations of the articular ranges being the instruments used in most situations. Although fast and simple, these instruments are either prone to biases based on patients' interpretation and psychological state or provide only a partial evaluation of the functional limitations. Considering that the main function of the upper limb (composed of the shoulder, arm, elbow and forearm) is to place and orient the hand in space, for the latter to perform a particular action; the reachable space defined as the total volume where the upper limb is able to place the hand, is a feature of great interest. In biomechanical research laboratories, measurements with standard motion capture system enables a more comprehensive and accurate evaluation of the whole upper limb, including the evaluation of its reachable space. Unfortunately, the use of such platforms is generally too complex, expensive and time-consuming for clinical examinations. To this end, the development of new tools able to bring some of the most interesting features recorded in biomechanical laboratories to the clinical context would offer promising new solutions for the functional evaluation of patients and represent the second goal of this thesis. This thesis proposes two original instruments for the evaluation and robotic simulation of the glenohumeral joint. It offers a tool for reproducing physiological forces and motions within the glenohumeral joint, opening new opportunities for the study of glenohumeral instability. At the same time, it also proposes a second instrument allowing to evaluate the functional impairment of a patient's upper limb in an objective and reliable way, extracting information that in the past required the deployment of unrealistic amounts of resources for a clinical examination.