Gait analysis in children with cerebral palsy: bridging the gap between the laboratory and real life

Like 17 million people worldwide, an individual with cerebral palsy (CP) does not have the opportunity to walk harmoniously in society due to long-life impairments in movement and posture. The natural course of CP can be modulated by treatments and therapies that are nowadays mostly decided on the basis of assessments performed in clinical settings. The Clinical Gait Analysis (CGA) consists of a set of instrumented assessments aiming to obtain precise and quantitative information about a patientâs gait deviations, in order to better identify his motor disorders and their possible causes. However, it is not clear whether gait assessments in clinics (âcapacityâ) are representative of daily-life behaviors (âperformanceâ). In this context, the present thesis aimed at exploring the gap between gait assessed in the laboratory and in real life for children with CP, as compared to children with typical development (TD). Two main objectives were settled: (i) to propose an objective and validated tool for gait assessments in a daily-life context with the highest possible accuracy as compared to clinical standard references; and (ii) to compare gait characteristics between both environments, the laboratory and the real life. Considering the immense progress in the design of wearable sensors, notably inertial measurement units (IMU), great enthusiasm recently arose for their use in ambulatory monitoring. IMUs, including 3D accelerometers and gyroscopes, were thus exploited in this work as a solution to measure gait features in the childrenâs daily life. A comparative study was first carried out to determine the most appropriate wearable system to be used for children with CP and for long-term measurements. Sensors located on the shanks and thighs and associated algorithms revealed to be the best solution. Next, a proof-of-concept study was completed and emphasized the need for personalized data processing for children with CP but also with TD. The second part (ii), dealing with the comparisons of walking capacity versus performance, adopted a progressive approach. The first step was the comparison between standardized walking, i.e. during a CGA protocol, and walking under challenging situations, such as dual tasks, i.e. thinking and talking while walking, in the laboratory. We found that dual tasks were responsible for lower motor gait capacities. The next step was the effective comparison of gait characteristics measured in the laboratory with the same gait characteristics measured in real-life settings, using the previously determined wearable system. First, walking speed, a global indicator of gait, and second, multiple gait parameters were compared between laboratory and daily life. Through two studies, we evidenced that children with CP have highly heterogenous behaviors but tend to perform better in clinical settings. Besides, we have highlighted that capacity is associated with performance in children with CP when they are evaluated with the same metrics. Through this doctoral work, the great challenges of using IMUs for gait analysis of children with CP have been highlighted. The proposed solutions reached a compromise between accuracy, number of outcomes, and acceptance. Furthermore, the presented clinical results proved with objective and quantitative evidence the existence of a gap between gait assessed in the laboratory and gait in real life, which could help clinicians to devise therapeutic plans better tailored to each childâs needs.

Assessment of Foot Signature Using Wearable Sensors for Clinical Gait Analysis and Real-Time Activity Recognition

Benoît Mariani

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.

EPFL2012

Instrumented shoes for daily activity monitoring in healthy and at risk populations

Christopher Moufawad El Achkar

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 âcomplexityâ, was discussed. Participants positively rated their comfort while using the system... (Please refer to thesis for full abstract)

EPFL2016

Gait in real world: validated algorithms for gait periods and speed estimation using a single wearable sensor

Abolfazl Soltani

Mobility concerns most daily tasks (e.g., householding, shopping), affecting life quality. Gait speed, recognized as the sixth vital sign, is a key to characterize mobility. It is also a primary outcome of many clinical interventions. Monitoring gait in unsupervised free-living situations is crucial. It offers the possibility to assess purposeful gait (e.g., catching a bus) in contextual situations (e.g., socializing), multitasking conditions requiring attention, and where the activity is affected by environmental components (e.g., buildings, streets). The Global Navigation Sattelite System (GNSS) measures real-world gait speed, but it suffers from high power consumption and is available only outdoors. Multiple Inertial Measurement Units (IMU, including accelerometer and gyroscope) worn on the body could be used to estimate speed accurately. However, it is challenging and cumbersome to wear them every day. Another alternative is to use a single IMU, where the wrist and the Lower Back (LB) are recognized as appropriate sensor locations for real-life conditions. The Wrist-mounted IMU could be integrated inside a watch, thus, increasing user satisfaction. The LB-worn IMU could capture robust gait patterns even for patients and could be used to extract gait parameters like asymmetry. The wrist-based algorithms are mostly validated in supervised situations. They significantly lose their performance in daily life. While many LB-based methods exist, they have not been fully compared to determine what algorithms and under what criteria (slow, normal, and fast walkers) lead to better performance. This thesis primarily presents accurate wrist-worn IMU-based (with barometer) speed (including cadence and step length) estimation and gait bout detection algorithms using Machine Learning (ML). An online personalization was devised in which the GNSS was sporadically used to capture a few speed data of a person's gait to tune the speed model gradually. Biomechanically-derived features were also extracted based on acceleration intensity, periodicity, noisiness, and wrist posture. The gait bout detection algorithm was validated against a multiple-IMU-based system for healthy people in unsupervised daily life. High sensitivity, specificity, accuracy were achieved (90, 97, 96 %). The personalized speed algorithm was also validated against GNSS for healthy subjects in real-world conditions, reaching an accuracy of 0.05 and 0.14 m/s for walking and running. Furthermore, this thesis performs cross-validation on the LB-based algorithms to investigate the best algorithms for different speed ranges. Twenty-nine algorithms were organized in a conceptual framework, improved, and implemented. A novel combination technique was also proposed. The cross-validation against an instrumented mat and a multiple-IMU-based algorithm on both healthy and patient populations offered the combined approach as an accurate and robust solution with an error of 0.10 m/s. Finally, this thesis demonstrates the feasibility of using the proposed wrist-based algorithms for long-duration monitoring of gait in a large cohort study (around 2800 subjects). Results showed that the gait speed significantly improves frailty and handgrip strength estimations. Overall, the proposed algorithms are independent of sensor orientation, thus, easy-to-use. A single IMU offers a high battery life and comfort, perfect for long-duration outdoors/indoors monitoring.

EPFL2020