Dissertations / Theses on the topic 'Gait'

Biometric technology is rapidly evolving in today's society. A large part of the technology has its roots hundreds, or even thousands of years back in time, while other parts are new and futuristic. Research suggest that individuals can be identified by the way they walk, and this kind of biometrics, gait biometrics, is a rather new and definitely intriguing field. However, the technology is far from mature; the performance is not generally competitive to other biometrics, and it has not been thoroughly tested security-wise. This thesis aims to test the security strength of gait biometrics. It will focus on imitation, or mimicking of gait. The bottom line question is whether it is possible to learn to walk like someone else. If this turns out to be easy, it will have a severe effect on the potential of gait as an authentication mechanism in the future. The report is logically twofold. In one part, the reader is brought up to speed on the field of gait biometrics, and a software tool for gait authentication is developed and presented. Second, an experiment is conducted, involving extensive training of test subjects, and using sources of feedback like video and statistical analysis. The data is analyzed by regression, and the goal is to determine whether or not the participants are increasing their mimicking skills, or simply put: if they are learning. The first part of the experiment involves 50 participants that are successfully enrolled using the developed software. The results compete with state of the art gait technology, with an EER of 6.2%. The rest of the experiment is related to mimicking, and the thesis discovers that six out of seven participants seem to have a natural boundary to their performance, a "plateau", forcing them back whenever they attempt to improve further. The location of this plateau predetermines the outcome of an attack; for success it has to lie below the acceptance threshold corresponding to the EER. Exactly one such boundary is identified for almost all participants, but some data also indicate that more than one plateau can exist simultaneously. The final result however, is that a very limited amount of learning is present, not nearly enough to pose a threat to gait biometrics. Gait mimicking is a hard task, and our physiology works against us when we try to adopt specific gait characteristics.

Rapid advances in biometrics technology makes their use for person‘s identity more acceptable in a variety of applications, especially in the areas of the interest in security and surveillance. The upsurge in terrorist attacks in the past few years has focused research on biometric systems that have the ability to identify individuals from a distance, and this is spearheading research interest in Gait biometric due to being unobtrusive and less dependent on high image/video quality. Gait biometric is a behavioral trait that aims to identify individuals from image sequences based on their walking style. The growing list of possible civil as well as security applications for various purposes is paralleled by the emergence of a variety of research challenges in dealing with a various external as well as internal factors influencing the performance of Gait Recognition (GR) in unconstrained recording conditions. This thesis is concerned with Gait Recognition in unconstrained scenarios aims to address research questions covering (1) The selection of sets of features for a gait signature; (2) The effects of gender and/or recoding condition case (neutral, carrying a bag, coat wearing) on the performance of GR schemes; (3) Integrating gender and/or case classifications into GR; and (4) The role of emerging Kinect sensor technology, with its capability of sensing human skeletal features in GR and applications. Accordingly, our objectives will focus on investigating, developing and testing the performance of using a variety of gait sequencefeatures for the various components/tasks and their integration. Our tests are based on large number of experiments based on CASIA B database as well as an in-house database of Kinect sensor recording. In all experiments, we use different dimension reduction and feature selection methods do reduce the dimensions in these proposed feature vectors, such as Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA) and Fisher Score, followed by different classification methods like; k-nearest-neighbour (k-NN), Support Vector Machine (SVM), Naive Bayes and linear discriminant classifier (LDC), to test the performance of the proposed methods. The initial part is focused on reviewing existing background removal for indoor and outdoor scenarios and developing more efficient versions primarily by adopting the work for wavelet domain rather than the traditional spatial domain based schemes. These include motion detection by frame differencing and Mixture of Gaussians, the latter being more reliable for outdoor scenarios. Subsequently, we investigated a variety of features that can be extractedfrom various subbands of wavelet-decomposed frames of different body parts (partitioned according to the golden ratio). We gradually built sets of features, together with their fused combinations, that can categorized as hybrid of model-based and motion-based models. The first list of features developed to deal with Neutral Gait Recognition (NGR) includes: Spatio-Temporal Model (STM), Legs Motion Detection Feature (LMD), and the Statistical model of the approximation LL-wavelet subband images (AWM). We shall demonstrate that fusing these features achieves accuracy of 97%, which is comparable to the state of the art. These features will be shown to achieve 96% accuracy in gender classification (GC), and we shall establish that the NGR2 scheme that integrates GC into NGR improves the accuracy by a noticeable percentage. Testing the performance of these NGR schemes in recognising non-neutral cases revealed the challenges of Unrestricted Gait Recognition (UGR). The second part of the thesis is focused on developing UGR schemes. For this, first a new statistical wavelet feature set extracted from high frequency subbands, called Detail coefficients Wavelet Model (DWM) was added to the previous list. Using different combinations of these schemes, will be shown to significantly improve the performance for non-neutral gait cases, but to less extent in the coat wearing case. We then develop a Gait Sequence Case Detection (GSCD) which has excellent performance. We will show that integrating GSCD and GC together into UGR improves the performance for all cases. We shall also investigate the different UGS scheme that generalizes existing work on Gait Energy and Gait Entropy images (GEI and GEnI) features but in the wavelet domain and in different body parts. Testing these two schemes, and their fusion, post the PCA dimension reduction yield much improved accuracy for the non-neutral cases compared to existing scheme GEI and GEnI schemes, but are significantly outperformed by the last scheme. However, by fusing the UGS scheme with the GSCD+GC+UGR scheme above we will get best accuracy that outperform the state of the art in GR specially in the non-neutral cases. The thesis ended by conducting a rather limited investigation on the use of the Kinect sensors for GR. We develop two sets of features: Horizontal Distance Features and Vertical Distance Features from small set of skeleton point trajectories. The experimental result on neutral was very successful but for the unrestricted gait recognition (with the 5 case variations) satisfactory but not optimal performance relies on the gallery including balanced number of samples from all cases.

Two methods were used in this research. The pedobarograph is a relatively novel method of gait analysis for animals which enables footfall patterns to be recorded, enabling spatial parameters (step length, width and angle) and plantar pressure patterns to be described and measured. A Kistler force plate was then used to measure the three-dimensional ground reaction forces (GRF's) produced during walking. Speed and cadence can be calculated using either system. Gait patterns are described for normal birds, and for different strains of broilers, raised on different feeding regimes. All the gait parameters were very variable, both between birds, and within the same bird, even when bodyweight and speed were controlled for. Despite the high variability, however, significant differences were identified in many of the gait parameters between the different groups. The vertical and craniocaudal GRF's of Brown Leghorns showed similar characteristics to those produced in human walking. The peak vertical forces were of a similar order of magnitude in the birds as in humans (125-150 % bodyweight), and the peak craniocaudal forces, and the rate of change of force, were closely tied to speed. All the GRF's in the birds increased significantly with increasing speed, except for braking rate (which was more variable) and stance time (which decreased significantly). The mediolateral forces were much greater in the birds than have been reported for other species, however, with peaks of 10-22% bodyweight. Analysis of plantar pressures showed that the pressure were concentrated on the digital pads, with the lowest pressure on the metatarsal pad (131 kNm-2), and highest pressure on the medial toe (up to 218 kNm-2).Combined gait analysis and morphometric studies of ad libitum-fed selected broilers identified many ways in which their gait deviated from that of relaxed broilers and Brown Leghorns, in ways which would serve to increase stability and decrease stresses on the skeleton.

This item is only available in print in the UCF Libraries. If this is your Honors Thesis, you can help us make it available online for use by researchers around the world by following the instructions on the distribution consent form at http://library.ucf
Bachelors
Engineering and Computer Science
Computer Science

This master’s thesis project “Instrumentation of Gait Analysis” was carried out at and funded by Integrum AB, Gothenburg, Sweden. Force analysis is critical during rehabilitation process of amputation patients, since overloading might place the bone-implant interface at risk; while underloading might extend unnecessarily the already long rehabilitation program [1]. Highly developed sensor and data acquisition technology provides an easy and reliable way to do force analysis. This thesis introduces the problem and provides background material regarding Orthotics and Prosthetics, including osseointegration. The existing gait analysis techniques and sensor technology will be described. Based upon the criteria that are introduced, a suitable sensor and integration platform was selected to implement a new gait analysis system. Several trials of different gait states are proposed using the prototype to do gait analysis, the results are presented and analyzed. The success of this prototype has lead to plans to design an Osseointegrated Prostheses for the Rehabilitation of Amputees(OPRA) product

The work presented within this document details the development of a novel gait verification system suitable for a variety of applications such as human motion studies, medical analysis and security situations. The human gait is a spatio temporal process involving the coordination and interaction between the nervous, skeletal and muscular systems. Due to inherent variations in the limb lengths, muscle strengths and body mass gait is inherently individual. To develop a suitable feature extraction process a virtual gait laboratory was developed. The virtual laboratory contains virtual character templates articulated with a 32 bone skeleton system using motion capture data. Data was extracted from the character as a series of X, Y and Z translations for pro cessing. The virtual laboratory allows the testing of data extraction processes without the need for direct testing on human subjects. Feature extraction was performed using Principal Component Analysis (PCA). PCA allows data to be compressed and describes as a series of principal scores (PC) containing the weightings of the data. Feature extraction was performed on human subjects with the motions applied to a skeletal system containing individual physical dimensions. A second set of features was created by applying the motions to a single skeletal system. This removed the interpersonal variations from the dataset to explore the difference in classification when these variables have been removed. Overall generic motions are present within the first PC score. Higher PC scores contain unique motion characteristics suitable for classification of the subject s within a database. To verify a subject within the database Linear Discriminant Analysis (LOA) was performed. LOA projects data as a linear combination of features using a t raining data set of known outcomes. A subsequent sample can then be projected into the linear space for classification and verification of the subject within the database.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2002.
Includes bibliographical references (p. 121-124).
This thesis describes a representation of gait appearance for the purpose of person identification and classification. This gait representation is based on simple localized image features such as moments extracted from orthogonal view video silhouettes of human walking motion. A suite of time-integration methods, spanning a range of coarseness of time aggregation and modeling of feature distributions, are applied to these image features to create a suite of gait sequence representations. Despite their simplicity, the resulting feature vectors contain enough information to perform well on human identification and gender classification tasks. We demonstrate the accuracy of recognition on gait video sequences collected over different days and times, and under varying lighting environments. Each of the integration methods are investigated for their advantages and disadvantages. An improved gait representation is built based on our experiences with the initial set of gait representations. In addition, we show gender classification results using our gait appearance features, the effect of our heuristic feature selection method, and the significance of individual features.
by Lily Lee.
Ph.D.

Human gait identification has become an active area of research due to increased security requirements. Human gait identification is a potential new tool for identifying individuals beyond traditional methods. The emergence of motion capture techniques provided a chance of high accuracy in identification because completely recorded gait information can be recorded compared with security cameras. The aim of this research was to build a practical method of gait identification and investigate the individual characteristics of gait. For this purpose, a gait identification approach was proposed, identification results were compared by different methods, and several studies about the individual characteristics of gait were performed. This research included the following: (1) a novel, effective set of gait features were proposed; (2) gait signatures were extracted by three different methods: statistical method, principal component analysis, and Fourier expansion method; (3) gait identification results were compared by these different methods; (4) two indicators were proposed to evaluate gait features for identification; (5) novel and clear definitions of gait phases and gait cycle were proposed; (6) gait features were investigated by gait phases; (7) principal component analysis and the fixing root method were used to elucidate which features were used to represent gait and why; (8) gait similarity was investigated; (9) gait attractiveness was investigated. This research proposed an efficient framework for identifying individuals from gait via a novel feature set based on 3D motion capture data. A novel evaluating method of gait signatures for identification was proposed. Three different gait signature extraction methods were applied and compared. The average identification rate was over 93%, with the best result close to 100%. This research also proposed a novel dividing method of gait phases, and the different appearances of gait features in eight gait phases were investigated. This research identified the similarities and asymmetric appearances between left body movement and right body movement in gait based on the proposed gait phase dividing method. This research also initiated an analysing method for gait features extraction by the fixing root method. A prediction model of gait attractiveness was built with reasonable accuracy by principal component analysis and linear regression of natural logarithm of parameters. A systematic relationship was observed between the motions of individual markers and the attractiveness ratings. The lower legs and feet were extracted as features of attractiveness by the fixing root method. As an extension of gait research, human seated motion was also investigated.

The purpose of the study was to identify the lower limb kinematic adaptations made in normal gait to accommodate to static transverse slopes. Five male subjects were asked to walk along a platform at 0%, 5% and 10% slope. Kinematic data for the ankle, knee and hip were collected at 60Hz using the Ultratrak RTM (Polhemus Inc., Burlington, VT, USA) electromagnetic tracking system. Results indicated that significant (p < 0.05) joint angle changes occurred in both the uphill (UH) and downhill (DH) lower limbs. The adaptations served as compensatory changes to functionally shorten the UH limb and lengthen the DH limb.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2004.
Includes bibliographical references (p. 216-220).
After a stroke, persons suffer from neurological impairments that affect gait, and so require rehabilitation to regain ambulatory function. While 82% of patients recover the ability to walk, current methods including physiotherapy and partial body-weight supported treadmill training (PBWSTT) are monotonous and require intense therapist effort. The Mechanized Gait Trainer and the LOKOMAT are two robotic devices that have been developed to improve gait rehabilitation, but neither provides the facilitation of pelvis movements afforded by traditional methods. In addition, neither device is truly backdrivable. As shown by Hogan and Krebs, backdrivable, impedance-controlled robots are ideal for rehabilitation because of their stable interaction properties. Robots for the arm/shoulder, wrist, ankle, and hand have already been developed. This thesis describes the design of a robot for gait rehabilitation through the facilitation of pelvis movements. Four degrees of freedom (DOF) are actuated: vertical, lateral, and frontal translations as well as the rotation about the vertical axis. Vertical forces support part of the patient's weight. Lateral forces assist the weight shift from stance leg to swing leg and are a part of physiotherapy and treadmill training.
(cont.) Frontal forces help pace the subject as on a treadmill. Pelvic rotations can impart energy into the swing leg without direct actuation of the hip and knee muscles. A four DOF mechanism was designed to control these movements, consisting of a three-DOF planar linkage with a vertical prismatic translation. A mockup of the configuration was designed and tested to show that the non-actuated pelvis DOFs are not adversely affected by the device. Design calculations include finding the optimal linkage configuration, selecting ballspline shafts for the vertical DOF, selecting actuators, and designing the robot arm cross-sections and joints. A final design for the four-DOF module is presented.
by Michael H. Roberts.
S.M.

Gait biometrics has attracted increasing interest in the computer vision and machine learning communities because of its unique advantages for recognition at distance. However, there have as yet been few gait biometric approaches which use temporal three-dimensional(3D) data. Clearly, 3D gait data conveys more information than 2D gait data and it is also the natural representation of human gait as perceived by humans.The University of Southampton has created a multi-biometric tunnel using twelve cameras to capture multiple gait images and reconstruct them into 3D volumetric gait data. Some analyses have been done using this 3D dataset mainly to solve the view dependent problem using model-free silhouette-based approaches. This thesis explores the potential of model-based methods in an indoor 3D volumetric gait dataset and presents a novel human gait features extraction algorithm based on marionette and mass-spring principles. We have developed two different model-based approaches to extract human gait kinematics from 3D volumetric gait data. The first approach used a structural model of a human. This model contained four articulated cylinders and four joints with two degrees of rotational freedom at each joint to model the human lower legs. Human gait kinematic trajectories were extracted by fitting the gait model to the gait data. We proposed a simple yet effective model-fitting algorithm using a correlation filter and dynamic programming. To increase the fitting performance, we utilized a genetic algorithm on top of this structural model. The second approach was a novel 3D model-based approach using a marionette-based mass-spring model. To model the articulated human body, we used a stick-figure model which emulates marionette's motion and joint structure. The stick-figure model had eleven nodes representing the human joints of head, torso and lower legs. Each node was linked with at least one other node by spring. The voxel data in the next frame had a role as an attractor which able to generate forces for each node and then iteratively warp the model into the data. This process was repeated for successive frames. Our methods can extract both structural and dynamic gait features. Some of the extracted features were inherently unique to 3D gait data such as footprint angle and pelvis rotation. Analysis on a database of 46 subjects shows an encouraging correct classification rate up to 95.1% and suggests that model-based 3D gait analysis can contribute even more in gait biometrics.

Human gait is a complex process that involves the coordination of the central nervous and muscular systems. A disruption to the either system results in the impairment of a person’s ability to walk. Impairments can be caused by neurological disorders such as stroke and physical conditions like amputation. There is not a standardized method to quantitatively assess the gait asymmetry of affected subjects. The purpose of this research is to understand the fundamental aspects of asymmetrical effects on the human body and improve rehabilitation techniques and devices. This research takes an interdisciplinary approach to address the limitations with current rehabilitation methodologies. The goal of my Doctoral research is to understand the fundamental effects of asymmetry caused by physical and neurological impairments. The methods discussed in this document help in developing better solutions to rehabilitate impaired individuals’ gait. I studied four major hypothesis in regards to gait asymmetry. The first hypothesis is the potential of asymmetric systems to have symmetric output. The second hypothesis is that a method that incorporates a wider range of gait parameter asymmetries can be used as a measure for gait rehabilitation. The third hypothesis is that individuals can visually identify subtle gait asymmetries. Final hypothesis is to establish the relationship between gait quality and function. Current approaches to rehabilitate impaired gait typically focus on achieving the same symmetric gait as an able-body person. This cannot work because an impaired person is inherently asymmetric and forcing them to walk symmetrically causes them to adopt patterns that are not beneficial long term. Instead, it is more prudent to embrace the asymmetry of the condition and work to minimize in specific gait parameters that may cause more harm over the long run. Combined gait asymmetry metric (CGAM) provides the necessary means to study the effect of the gait parameters and it is weighted to balance each parameter’s effect equally by normalizing the data. CGAM provides the necessary means to study the effect of the gait parameters and is weighted towards parameters that are more asymmetric. The metric is also designed to combine spatial, temporal, kinematic, and kinetic gait parameter asymmetries. It can also combine subsets of the different gait parameters to provide a more thorough analysis. CGAM will help define quantitative thresholds for achievable balanced overall gait asymmetry. The studies in this dissertation conducted on able-body and impaired subjects provides better understanding of some fundamental aspects of asymmetry in human gait. Able body subjects test devices that aim to make an individual’s gait more asymmetric. These perturbations include a prosthetic and stroke simulator, addition of distal mass, and leg length alterations. Six able-body subjects and one amputee participated in the experiment that studied the effect of asymmetric knee height. The results which consisted of analyses of individual gait parameters and CGAM scores revealed that there is evidence of overall reduction of asymmetry in gait for both able-body subject on prosthetic simulators and transfemoral amputee. The transfemoral amputee also walked with a combination of distal mass with lowered knee height. Although this configuration showed better symmetry, the configuration is detrimental in terms of energy costs. Analyzing the data of gait with the stroke simulator showed that the subject’s gait does undergo alterations in terms of overall gait asymmetry. The distal mass and leg length alteration study has revealed some significant findings that are also reflected in the prosthetic study with distal mass. A leg length discrepancy (LLD) or the change of limb mass can result in asymmetric gait patterns. Although adding mass and LLD have been studied separately, this research studies how gait patterns change as a result of asymmetrically altering both leg length and mass at a leg’s distal end. Spatio-temporal and kinetic gait measures are used to study the combined asymmetric effects of placing LLD and mass on the opposite and same side. There were statistically significant differences for the amount of mass and leg length added for all five parameters. When LLD is added to longer leg, the temporal and kinetic gait parameters of the shorter limb and the altered limb’s spatial parameter become more asymmetric. Contrary to the hypothesis, there was no significant interaction between the amount of mass and leg length added. There were cases in all perturbations where a combination of mass and LLD make a gait parameter more symmetric than a single effect. These cases exhibit the potential for configurations with lower overall asymmetries even though each parameter has a slight asymmetry as opposed to driving one parameter to symmetry and other parameters to a larger asymmetry. CGAM analysis of the results revealed that the addition of distal mass contributes more towards overall asymmetry than LLD. Analyzing 11 gait parameters for LLD and mass on the same side showed that the overall asymmetry decreased for the combination of small LLD and mass. This is consistent with the findings from analyzing five individual gait parameters. Impaired subjects include individuals with stroke and amputees. The clinical trials for individuals with stroke involve training with the Gait Enhancing Mobile Shoe (GEMS) that pro- vides an asymmetric effect on the subject’s step length and time. Training with the GEMS showed improvement in clinical measures such as timed up and go (TUG), six minute walk test (6MWT), and gait velocity. The subjects also showed lower step length symmetry as intended by the GEMS. The ground reaction force asymmetries became more asymmetric as the spatial and temporal parameters became more symmetric. This phenomenon shows evidence that when an individual with stroke is corrected, for spatial and temporal symmetry is at the expense of kinetic symmetry. The CGAM scores also reflected similar trends to that of spatial and temporal symmetry and the r2 correlation with the gait parameters proved that double limb support asymmetry has no correlation with CGAM while ground reaction force asymmetry has a weak correlation. Step length, step, and swing time showed high correlation to CGAM. I also found the r2 correlation between the clinical measures and the CGAM scores. The CGAM scores were moderately correlated to 6MWT and gait velocity but had a weak correlation with TUG. CGAM has positive correlation with TUG and has negative correlation with 6MWT and gait velocity. This gives some validation to CGAM as a potential metric that can be used to evaluate gait patterns based on their asymmetries. Transfemoral amputees were tested for their gait with varied prosthetic knee heights to study the asymmetrical effects and trained split-belt treadmill. Asymmetric knee heights showed improvement in multiple gait parameters such as step length, vertical, propulsive, and braking force asymmetry. It also decreased hip and ankle angle asymmetries. However, these improvements did lead other parameters to become more asymmetric. The CGAM scores reflect this and they show overall improvement. Although the lowest knee height showed improvement, the input from the amputee suggested that the quality of gait decreased with the lowest knee height. These exploratory results did show that a slightly lower knee height may not affect the quality of gait but may provide better overall symmetry. Another exploratory study with split-belt treadmill training, similar to the protocol followed for individuals with stroke, showed definitive improvement in double limb support, swing time, step length and time symmetry. This was also reflected in the improvements seem post training in the CGAM scores as well. I found the r2 correlation of the CGAM and the gait parameters including gait velocity. Step length and swing time show consistent correlation for individual subjects and all the data combined to CGAM. Gait velocity shows a moderate correlation to CGAM for one subject and a high correlation to the other one. However, the combined data of gait velocities does not have any correlation with CGAM. These results show that CGAM can successfully represent the overall gait parameter asymmetry. The trends seen in the gait parameters is closely reflected in the CGAM scores. This research combines the study of asymmetry with people’s perception of human gait asymmetry, which will help in estimating the thresholds for perceivable asymmetrical changes to gait. Sixteen videos were generated using motion capture data and Unity game engine. The videos were chosen to represent the largest variation of gait asymmetries. Some videos were also chosen based on CGAM values that were similar but had large variation in underlying gait parameters. The dataset consisted of results of perturbation experiments on able-body subjects and asymmetric knee height prosthesis on transfemoral amputee. These videos were rated on a seven point Likert scale by subjects from 7 being normal to 1 being abnormal. Thirty one subjects took part in the experiment, out of which only 22 subject’s data was used because they rated at least 3 videos. The results show that the subjects were able to differentiate asymmetric gait with perturbations to able-body gait without perturbation at a self-selected speed. r2 correlation analysis showed that hip angle had mild correlation to the Likert scale rating of the 16 different gait patterns. Multivariate linear regression analysis with a linear model showed significant contribution of ankle and hip angles, vertical, propulsive, and braking forces. It is interesting that the majority of parameters that showed significance are not perceivable visually. Ankle and hip angles are visually perceivable and this significance revealed that subjects seemed to perceive asymmetric ankle and hip angles as abnormal. However, the subjects do not perceive asymmetric knee angles as completely abnormal with evidence of no significance, no correlation, and neutral Likert rating for gait patterns that perturbed knee angles.

This research has advanced our understanding of how the brain controls muscle activity during walking, in healthy adults of different ages and people with Parkinson's disease. It has shown that there is direct brain control of leg muscles during the double support phase of gait in these populations. Moreover, this brain control is reduced in people with Parkinson's compared to healthy individuals, and standard anti-Parkinsonian medication does not counteract this deficiency.

Tidigare forskning har funnit samband mellan gång och kognitiv funktion, men sambandet mellan lindrig kognitiv svikt (MCI) och gång är inte helt klarlagt. Syftet med den föreliggande examensuppsatsen var således att undersöka hur gångvariabler och gångvariabilitet är påverkat hos personer med MCI i relation till kognitivt friska individer. Studien genomfördes i det pågående projektet the Healthy Ageing Initiative vid Umeå Universitet. Totalt 1937 personer inkluderades i studien. Samtliga var 70 år gamla och bosatta i Umeå kommun. Totalt 112 personer bedömdes ha MCI utifrån instrumentet Mini-Mental State Examination (MMSE). Gånganalys genomfördes med den elektroniska gångmattan GAITRite® system och deltagarna utförde fyra gångförsök: egenvald hastighet, snabb hastighet, kognitiv dual task och motorisk dual task. Resultatet visade att det finns skillnader mellan grupper vad gäller både spatiala och temporala aspekter, främst i de tre första försöken. Exempelvis uppvisade gruppen MCI lägre gånghastighet, kortare steg och kliv samt längre double support och kortare swing. Gruppen MCI uppvisade högre variabilitet under kognitiv dual task. Ett flertal gång- och variabilitets-variabler under kognitiv dual task kunde, enligt logistisk regression, predicera sannolikheten att ha MCI. Resultaten indikerar att gången hos de med MCI kan ge ökad fallrisk.
Previous research has found a connection between gait and cognitive function. However, the relationship between mild cognitive impairment (MCI) and gait has not been fully explored. Thus, the aim of this study was to examine how spatiotemporal gait parameters, and gait variability, are affected in people with MCI compared to cognitively healthy individuals (CHI). The study was carried out in cooperation with the Healthy Ageing Initiative research project, Umeå University, Sweden. A total of 1937 participants were included in the study. All participants were 70-years old and residents of the municipality Umeå. A total of 112 participants were classified as having MCI, as measured with the Mini-Mental State Examination (MMSE). Gait analysis was performed with the GAITRite® system, and participants performed four trials: preferred pace, fast pace, cognitive dual task and motor dual task. Results showed group differences in both spatial and temporal aspects of gait, especially during the first three trials. For example, participants with MCI walked more slowly, had shorter steps and strides, as well as a longer duration of the double support phase and lower duration of the swing phase. Participants with MCI revealed higher gait variability during cognitive dual task. Several of these variability variables, as well as spatiotemporal variables, could predict probability of having MCI, as seen through logistic regression. Results indicate that the gait observed in MCI could be related to a higher risk of falling.

In this thesis work an intelligent controller based on a gait synthesizer for a hexapod robot used in rescue operations is developed. The gait synthesizer draws decisions from insect-inspired gait patterns to the changing needs of the terrain and that of rescue. It is composed of three modules responsible for selecting a new gait, evaluating the current gait, and modifying the recommended gait according to the internal reinforcements of past time steps. A Fuzzy Logic Controller is implemented in selecting the new gaits.

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

This thesis analyzes the human action recognition problem. Human actions are modeled as a time evolving temporal texture. Gabor filters, which are proved to be a robust 2D texture representation tool by detecting spatial points with high variation, is extended to 3D domain to capture motion texture features. A well known filtering algorithm and a recent unsupervised clustering algorithm, the Genetic Chromodynamics, are combined to select salient spatio-temporal features of the temporal texture and to segment the activity sequence into temporal texture primitives. Each activity sequence is represented as a composition of temporal texture primitives with its salient spatio-temporal features, which are also the symbols of our codebook. To overcome temporal variation between different performances of the same action, a Profile Hidden Markov Model is applied with Viterbi Path Counting (ensemble training). Not only parameters and structure but also codebook is learned during training.

Gait initiation is a temporary movement between upright posture and steady-state gait. The activation of several postural muscles has been identified to precede changes observed in vertical reaction force. Previous research examining gait initiation has concentrated on the electromyographic activity of muscles of the lower limbs. Few studies, however, have looked at recruitment patterns of the muscles of the thigh and trunk. This study was conducted to determine the recruitment patterns and the roles of certain muscles of the trail and lead lower limbs and trunk for the duration from quiet stance to trail leg toe-off. Eleven healthy participants initiated gait with their right leg. Electromyographic data were collected bilaterally from the erector spinae, tensor fasciae latae, adductor magnus and tibialis anterior muscles. In addition, force platform data were recorded for the duration of quiet stance to toe-off of the trail limb. (Abstract shortened by UMI.)

Parkinson's disease is a degenerative neurological disease that often impacts older adults, leading to difficulties with transfers, posture, balance, and walking. Approximately half of those with Parkinson's disease develop the symptom of freezing of gait, a condition that makes initiating walking difficult. This leads to loss of independence, fear of falls, injuries, inactivity resulting in social isolation, and increased risk of osteoporosis. Pharmacological therapy provides relief to only some of the neurological symptoms in Parkinson's disease. However, gait, posture, balance, and freezing of gait obtain little benefit from drug treatments.

Research suggests visual, auditory, and vibratory cueing methods may improve some features of gait. However, visual cueing techniques often require modification of the environment, which is not possible in the community setting. Auditory cueing techniques can be executed with the use of headphones, but this can quickly become a problem, as it becomes a safety concern when it interferes with hearing environmental sounds. Studies of whole body vibration and studies of segment level vibration have demonstrated improvements to walking speed, step duration, step length, and gait variability. However, most studies have only focused on the immediate effects of vibration therapy.

To assess the impact of vibration as an intervention, we have designed the hard- ware, software, therapy protocol, and conducted a series of studies for this dissertation. The hardware is designed as a segment level device, which can be worn by a user and both provides the vibratory stimulation and measures parameters of gait. We call the device the PDShoe. Three force sensors are embedded into the insole the shoe worn by subjects. Two tactor actuators are placed between the shoe and subject's skin. Upon heel contact with the floor, a tactor placed at the heel begins to vibrate at a predetermined frequency and amplitude. Likewise, when the lateral metatarsal head or hallux of the foot contact the floor, a tactor placed over the metatarsal heads begins to vibrate. Vibration is only provided with the foot is in contact with the floor; this is step synchronized vibration. Data is transferred via a wireless network in real time to a host computer where it is stored for later analysis. Parameters of the vibratory stimulation can be set on the PDShoe from the host computer. Data analysis is performed in a numerical analysis environment, where the time series force data is cut into individual steps. Eight parameters of gait are extracted from each of these steps, providing mean and variability measures for each subject.

To test the efficacy of step synchronized vibration, we conducted two studies with healthy subjects and four studies with Parkinson's disease subjects. Accuracy and reliability of the device was established and included the use of change point analysis and correlation with the 10-meter Walk Test. Using the PDShoe system without vibratory feedback, we successfully characterized the gait of healthy persons over fifty years old. Subsequently we tested the feasibility, safety, and impact of our step synchronized vibration protocol with healthy subjects. Our protocol was easily implemented, well tolerated, and there were no adverse events. As was expected, no impact from the step synchronized vibration was seen with the healthy subjects. We then performed a feasibility and impact study with Parkinson's disease subjects. Again the protocol was easily implemented, well tolerated, and no adverse events were noted. A difference between pre- and post-intervention clinical scores of the Freezing of Gait Questionnaire, Berg Balance Scale, Timed Up and Go and gait measures of step and stance duration were found. Next, we conducted a clinical study with eight subjects at the All India Institute of Medical Sciences. This study demonstrated that subjects who exhibited the symptom of freezing of gait received the greatest benefit from the step synchronized vibration. These results informed a follow up study on seventeen subjects, all of whom had the symptom of freezing of gait. This last study demonstrated positive improvements in clinical measures of the Freezing of Gait Questionnaire, Berg Balance Scale, Timed Up and Go, Walking Speed, and gait characteristics of step duration, stance duration, swing duration, heel max force timing, and heel contact duration. There is encouraging evidence for further investigation; this data will be used to inform future studies.

These studies demonstrated the functionality, reliability, validity of the PDShoe device to measure characteristics of gait. Impact was demonstrated in three studies with Parkinson's disease subjects.

Biometric technology has emerged as a viable identification and authentication solution with various systems in operation worldwide. The technology uses various modalities, including fingerprint, face, iris, palmprint, speech, and gait. Biometric recognition often involves images or videos and other image impressions that are fragile and include subtle details that are difficult to see or capture. Thus, there is a need for developing imaging applications that allow for accurate feature extraction from images for identification and recognition purposes. Biometric modalities can be classified into two classes: physiological (i.e. fingerprint, iris, face, palm-print) or behavioural traits (speech, gait). This work is concerned with an investigation of biometric recognition at a distance and the gait modality has been chosen for various reasons. Gait data can be captured at a distance and is non-invasive. Additionally, it has advantages such as the fact that a person’s gait is hard to copy, and by trying to do so, the imitator will likely appear more suspicious. Although, due to covariates, for example, a change in viewing angle, clothes, shoes, shadow or elapsed time can make gait recognition additionally challenging. There are several approaches for studying gait recognition systems such as model-based and model-free. This thesis is based on a model-free approach and proposes a supervised feature extraction approach capable of selecting distinctive features for the recognition of human gait under clothing and carrying conditions. In this work; to allow for the characterisation of human gait properties for individual recognition, a spatiotemporal gait representation technique called Gait Energy Image (GEI) has been used. This approach is aimed at improving the recognition performance based on the principles of feature texture descriptors extracted from GEI. Furthermore, as part of this work, the dynamic parts of the energy gait representation have been proposed as means to extract more discriminative information from a gait sequence using reduction techniques in order to further improve the human identification rate. The four methods proposed were evaluated using CASIA Gait Database (dataset B) and USF Database under variations of clothing and carrying conditions for different viewing angles. The first method is based on Haralick texture feature, and use the RELIEF selection algorithm. This method showed that a judicious deployment of horizontal GEI features outperforms similar methods by up to 7.00%. In addition, this method achieved an improved classification rate of up to 80.00% from a side view of 90o. The second and third contributions are concerned with an investigation of the Gabor filter bank and Multi-scale Local Binary Pattern (MLBP) as an efficient feature extraction for gait recognition under clothing distortions. To achieve this, various dimension reduction techniques including Kernel Principal Component Analysis, Maximum Margin Projection, Spectral Regression Kernel Discriminant Analysis and Locality Preserving Projections were investigated. The results showed that the proposed methods outperform the state-of-the-art counterparts by achieving up to 93.00% Identification Rate (IR) at rank-1 using the Gabor filter method, and achieving up to 92.00% IR using the MLBP method, when using a k-NN classifier for a side view of 90o. The final contribution of this work is concerned with an investigation of the Haar wavelet transform and its use for extracting powerful features for human gait recognition under clothing distortions. The experimental results using a k-NN classifier yielded attractive results of up to 93.00% in terms of highest IR at rank-1, compared to existing and similar state-of-the-art methods. It should be noted that all the experiments were carried out using the MATLAB programming environment.

Background: Stroke survivors fall more often, mostly due to stumbling and slipping; which may signify. These causes of falls are hypothesized to be caused by difficulty in controlling and adjusting foot placement in response to the environment. In healthy adults’ foot placement control is known to be influenced by balance control, available response time and executive function. All these factors are known to be affected by stroke; however, how these factors affect foot placement accuracy in stroke survivors is largely unknown. The overarching aim of this thesis is therefore to understand the role of these factors in the control of foot placement following stroke and by extension to better understand how foot-placement is controlled, the causes of stroke related impairments and potential reasons for falls. Methods/Results: Young (n=14) and older healthy adults (n=9) and stroke survivors (n=13) completed a series of experiments on a C-Mill (a force instrumented treadmill with visual projection of stepping targets) designed to assess the role of balance (study 1), response time (study 2) and executive function (study 3) on foot placement control in stroke survivors. Study 1 compared foot placement control in supported versus unsupported conditions; balance support reduced overall error while target stepping (main effect F (1,30)=18.141, p < 0.001), but mostly in stroke survivors. Study 2 compared foot placement control when targets could be seen in advance (planned) with targets appearing at midstance (reactive). Foot-placement error altered according to direction of step but not available response time, with significant increase in error (F (1,28) =6.013, p=0.021) when adjusting steps medio-laterally but decreased when adjusting steps anterio-posterially (F (1,28) = 5.932, p=0.021). Overall, stroke survivors missed about 10% of targets and undershot all targets while young healthy adults undershoot only lengthening steps. Study 3 evaluated the use of functional near-infrared spectrometry (fNIRS) to measure activation of prefrontal cortex activation (brain networks responsible for executive function) in target stepping conditions which can be expected to increase challenge to executive function. fNIRS showed high inter person variability and no systematic trends according to walking conditions. Conclusion: Stroke survivors miss about 1 in every 10 targets; in the real world this may lead to a fall. Balance support may generally help stroke survivors control foot-placement more accurately. However, the lack of difference in accuracy between reactive and pre-planned stepping indicates stroke survivors may respond to all foot-placement adaptations reactively (a “cluttered terrain strategy”). This ‘cluttered terrain strategy’ is indicative of increased cognitive control, however the use of fNIRS needs development to robustly be assess this during walking.

Thesis: S.M., Massachusetts Institute of Technology, School of Architecture and Planning, Program in Media Arts and Sciences, 2016.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 72-73).
In this thesis we present a novel approach to the computer simulation of forward dynamic gait models and the optimization of parameters that must be tuned for such models. This methodology is not limited to gait simulation, and could be useful for any situation in which a complex Simulink model requires variables to be tuned via machine learning to optimize all heuristic that can only be evaluated via simulation. Through the lens of Biomechatronic engineering research, we combine the fundamentals of software engineering with a refinement of the best practices of Matlab and Simulink programming and a working knowledge of inherent Matlab and Simulink constraints to construct a framework for rapid model development. Key features of this methodology include: the use of Simulink as an environment for rapidly prototyped models, the use of and construction of custom Simulink libraries, and use of the Matlab Optimization Toolbox. This methodology uses parallel evaluation of rapid acceleration Simulink executables to minimize optimization time, and allow research teams to take advantage of parallel processing and cloud computing. This methodology was applied to a bouncing gait model developed by Hartmut Geyer for evaluation. We demonstrate its effectiveness by simulating this model using a custom library of model components, such as ground contact model, Stateflow control, heuristic computation, and body segments.
by Matthew D. Furtney.
S.M.

Background: Gait analysis is used to measure objectively the ability to walk before and after hip and knee joint replacement. Most gait literature to date has measured patients walking at slow speeds, which may have unintentionally characterized successful lower limb operations as the same regardless of its operation type. The aim of this thesis was to use an instrumented treadmill to (1) test the upper limits of gait performance of patients awaiting and after lower limb arthroplasty and (2) describe and compare different joint procedures to healthy controls. Method: Gait analysis was carried out using an instrumented treadmill. Study participants walked at their preferred walking and top walking speed on the treadmill. Their vertical ground reaction force and spatiotemporal data was captured for both limbs by tandem force plates beneath the treadmill's belt. Results: The results suggest that at preferred walking speed on the flat, all subjects' gait characteristics were similar irrespective of joint state. The differences between patient groups became more apparent at top walking speed. Weight acceptance, step length and stride length were the most useful variables assessing these differences. Unicompartmental knee replacement and hip resurfacing had closer to normal gait patterns when compared to total knee replacement and conventional total hip replacement respectively. Predictably patients with knee osteoarthritis had asymmetrical gait patterns, impulse was found to be the best variable distinguishing patients with knee osteoarthritis from healthy controls. The downhill walking assessment established that patients with unicompartmental knee replacements had a more normal gait pattern than total knee replacement patients. Conclusion: The gait results suggest that an instrumented treadmill is a useful metric to describe patient gait patterns. Testing performance at higher walking speeds allowed differences to be detected, which were undetectable at slower speeds. Anatomically conserving procedures such as unicompartmental knee replacement and hip resurfacing appear to have convincing functional advantages compared to less conserving joint replacement procedures.

One of the unique advantages of human gait is that it can be perceived from a distance. A varied range of research has been undertaken within the field of gait recognition. However, in almost all circumstances subjects have been constrained to walk fronto-parallel to the camera with a single walking speed. In this thesis we show that gait has sufficient properties that allows us to exploit the structure of articulated leg motion within single view sequences, in order to remove the unknown subject pose and reconstruct the underlying gait signature, with no prior knowledge of the camera calibration. Articulated leg motion is approximately planar, since almost all of the perceived motion is contained within a single limb swing plane. The variation of motion out of this plane is subtle and negligible in comparison to this major plane of motion. Subsequently, we can model human motion by employing a cardboard person assumption. A subject's body and leg segments may be represented by repeating spatio-temporal motion patterns within a set of bilaterally symmetric limb planes. The static features of gait are defined as quantities that remain invariant over the full range of walking motions. In total, we have identified nine static features of articulated leg motion, corresponding to the fronto-parallel view of gait, that remain invariant to the differences in the mode of subject motion. These features are hypothetically unique to each individual, thus can be used as suitable parameters for biometric identification. We develop a stratified approach to linear trajectory gait reconstruction that uses the rigid bone lengths of planar articulated leg motion in order to reconstruct the fronto-parallel view of gait. Furthermore, subject motion commonly occurs within a fixed ground plane and is imaged by a static camera. In general, people tend to walk in straight lines with constant velocity. Imaged gait can then be split piecewise into natural segments of linear motion. If two or more sufficiently different imaged trajectories are available then the calibration of the camera can be determined. Subsequently, the total pattern of gait motion can be globally parameterised for all subjects within an image sequence. We present the details of a sparse method that computes the maximum likelihood estimate of this set of parameters, then conclude with a reconstruction error analysis corresponding to an example image sequence of subject motion.

Gait recognition algorithms are being increasingly widely researched, however a common assumption is that the subject will be presented side on to the camera. In practice it may not be possible to capture data from this view, so a useful gait recognition algorithm will have to provide a measure of orientation independence. Three gait recognition algorithms are examined and found to perform poorly with nonnormal orientation. The complex detail used for recognition can not be translated between orientations in a holistic silhouette manner. It is shown that orientation independent features can be extracted using a human model. The algorithm is developed and tested on live captured data and found to perform better across orientations than silhouette based approaches. The performance recorded at a single orientation is lower than that of other approaches, however only the motion of the subject is currently used for recognition. More accurate motion estimation will increase performance as will the inclusion of other model based features.

This thesis explores the dynamics of two bipedal, passive-walker models that are free to move in a three-dimensional environment. Specifically, two rigid-bodied walkers that can sustain anthropomorphic gait down an inclined plane with gravity being the only source of energy were studied using standard dynamical systems methods. This includes calculating the stability of periodic orbits and varying the system parameter to create bifurcation diagrams and to address the persistence of a periodic solution under specific parameter variations. These periodic orbits are found by implementing the Newton-Raphson root solving scheme. The dynamical systems associated with these periodic orbits are not completely smooth. Instead, they include discontinuities, such as those produced due to forces at foot contact points and during knee hyper-extension. These discontinuities are addressed in the stability calculations through appropriate discontinuity mappings. The difference between the two walker models is the number of degrees of freedom (DOF) at the hip. Humans possess three DOF at each hip joint, one DOF at each knee joint, and at least two DOF at each ankle joint. The first walker model studied had revolute joints at the hips and knees and completely locked ankles. To make the walking motion more anthropomorphic, additional degrees of freedom were added to the hip. Specifically, the second walker model has ball joints at the hips. Two control algorithms are used for controlling the local stability of periodic motions for both walker models. The methods, reference and delay feedback control, rely on the presence of discontinuities in the system. Moreover, it is possible to predict the effects of the control strategy based entirely on information from the uncontrolled system. Control is applied to both passive walker models to try and stabilize an unstable periodic gait by making small, discrete, changes in the foot orientation during gait. Results show that both methods are successful in stabilizing an unstable walking motion for a 3D model with one DOF in each hip and to reduce the instability of the walking motions for the model having more mobility in the hip joints.
Master of Science

A microcomputer-based video vector system has been developed to display the resultant foot-ground reaction force vector on a television image of the subject in real-time. The system consists of a strain gauged force platform, 12-bit ADC, a video camera, and an Archimedes microcomputer. The output of the computer is synchronised and overlaid on the video image of the camera by a video GENLOCK board. For each television field the resultant force vector is correctly positioned relative to the image of the force platform and superimposed on the TV image at 50 times/sec. A television/computer interface has been designed to detect the position of markers using the real-time video signals. Passive optical markers are used on the subject as a means of identifying regions of interst and a glass plate/light arrangement is used to enhance reflection signals from these markers. A threshold detector extracts marker peaks from the rest of the video and digital counters determine the spatial coordinates of the markers. A hardware video window selector has been designed which is selected prior to data collection to exclude the irrelevant areas in the video field. A 16-bit input/output system has been designed dedicated to data transfer between the TV/computer interface and the Archimedes microcomputer. The system has been tested in all of its aspects to verify its ability to produce repeatable and accurate results. Some of the typical results obtained from the system are presented for both normal and pathological gait. The aim to produce a simple to use and flexible instrument was maintained throughout the design phase. The result is an instrument which can be used to record the various useful gait parameters and be a valuable tool in routine gait evaluation in the clinic.

Intelligent devices such as smart phones, smart watches, virtual reality (VR) headsets and personal exercise devices have become integral elements of accessories used by many people. The ability of these devices to verify or identify the user could be applied for enhanced security and user experience customization among other things. Almost all these devices have built-in inertial sensors such as accelerometer and gyroscope. These inertial sensors respond to the movements made by the user while performing day to day activities like walking, getting up and sitting down. The response depends on the activity being performed and thus can be used for activity recognition. The response also captures the user's unique way of doing the activity and can be used as a behavioral biometric for verification or identification. The acceleration (accelerometer) and rate of rotation (gyroscope) are recorded in the device coordinate frame. But to determine the user's motion, these need to be converted to a coordinate frame relative to the user. In most situations the orientation of the device relative to the user can neither be controlled nor determined reliably. The solution to this problem requires methods to remove the dependence on device orientation while comparing the signals collected at different times. In a vast of majority of research to date, the performance of authentication algorithms using inertial sensors have been evaluated on small datasets with few tens of subjects, collected under controlled placement of the sensors. Very often stand alone inertial sensors have been used to collect the data. Stand alone sensors afford better calibration, while the sensors built into smart devices offer little or no means of calibration. Due to these limitations of the datasets used, it is difficult to extend the results from these research to realistic performance with a large number subjects and natural placement of off-the-shelf smart devices. This dissertation describes the Kabsch algorithm which is used to achieve orientation invariance of the recorded inertial data, enabling better authentication independent of device orientation. It also presents the Vector Cross Product (VCP) method developed to achieve orientation invariance. Details of a realistic inertial dataset (USF-PDA dataset) collected with commercial smart phones placed in natural positions and orientations using 101 subjects are given. The data includes sessions from different days on a subset of 56 subjects. This would enable realistic evaluation of authentication algorithms. This dataset has been made publicly available. The performance of five methods that address the orientation dependence of signals are compared to a baseline that performs no compensation for orientation of the device. The methods as a part of a overall gait authentication algorithm are evaluated on the USF-PDA dataset mentioned above and another large dataset with more than 400 subjects. The results show that the orientation compensation methods are able to improve the authentication performance on data with uncontrolled orientation to be close to performance on data collected with controlled orientation. The Kabsch method shows the highest improvement.

The effects of postural threat on gait initiation and steady state gait among Parkinson’s disease (PD) patients and age-matched adults were examined. Ten healthy adults (CTRL; mean age= 68.8 ± 8.4, range 56-80 years) and ten PD patients (PDOFF / ON ; mean age= 69.7 ± 10.3, range 54-81 years) initiated gait and continued with steady state walking along a walkway of two different height conditions. PD patients were first tested in a non-medicated state followed by testing in a medicated state. The results showed that gait initiation and steady state gait deficits inherent to PD are exacerbated in a postural threatening environment. As well, medication efficacy for overcoming parkinsonian deficits may be context dependent. These findings confirm the dynamic nature of movement deficits characteristic of parkinsonian patients and provide empirical evidence for specific environments that can create movement difficulties for people with PD.
x, 59 leaves : ill. ; 29 cm. --

In response to the potential benefits of task specific training in rehabilitation of gait after stroke and the need for affordable, simple ways to implement it, our group designed the elliptically based robotic gait trainer (EBRGT). A design review of the EBRGT, covering the design goals, an overview of the mechanical and electrical design, and a discussion of the novelty of the device and why it may be beneficial for individuals with hemiparesis secondary to stroke is discussed (Chapter 2). To characterize the new device, a study was performed to determine if the EBRGT produced a gait pattern that mimicked level surface walking in healthy adults (chapter 3). Sagittal plane kinematic analysis suggested the EBRGT produced joint movement patterns that are similar to level surface walking at the hip and knee with less similarity between activities at the ankle. Electromyography (EMG) revealed that the EBRGT induced a cyclic muscle firing pattern that had some similarities when compared to level surface walking. We also examined the feasibility of ambulatory individuals after stroke to use the EBRGT and if their movement patterns were similar to healthy adults walking on the same device (Chapter 4). All six participants were able to walk on the device with minimal assistance. These participants had joint kinematics and EMG similar to healthy adults, suggesting that individuals with hemiparesis perform a gait like movement when using the EBRGT. Lastly, a study was performed to determine if the EBRGT could improve gait parameters and function in ambulatory individuals with hemiparesis after stroke (chapter 5). Four participants walked on the EBRGT 3x/week for 4 or 8 weeks. After the intervention, all 4 participants increased their preferred gait speed. One participant had an improvement in gait speed that indicated functional gains. The results of this research suggest that the EBRGT can produce a gait pattern that has some similarities to level surface walking and that it is feasible for ambulatory individuals with hemiparesis to use the device. The device may also improve gait parameters in ambulatory individuals after stroke, but future studies with a control group need to be performed.

Si les neurosciences connaissent d’importants progrès dans l’imagerie et le génotypage, le phénotypage repose encore largement sur des échelles visuelles. Le phénotype chez l’homme repose principalement sur son style perceptivo-moteur qui donne une empreinte à la marche, la posture, l’équilibre, l’habilité des membres supérieurs, les mouvements oculaires etc. La marche, fonction complexe et fondamentale de l’être humain, implique l’ensemble du système musculo-squelettique, le système nerveux central et périphérique ainsi que les organes sensoriels. Elle est le produit d’un patron de marche automatique et inconscient modulé par le tronc cérébral, les noyaux gris centraux et par des retours sensitifs (visuels, proprioceptives, vestibulaires et épicritiques). Enfin, la marche est aussi sous contrôle volontaire. Le phénotypage quantitatif de la marche suppose la construction préalable de bases de données de signaux de marche d’un nombre élevé (centaines) de sujets et de patients. Ceci peut être mené à bien grâce à des outils de mesure simples d’utilisation et adaptés à la pratique médicale de routine. Il existe plusieurs moyens pour phénotyper la marche mais le capteur inertiel, en raison de son prix, de sa souplesse d’utilisation et de l’accès aux données brutes est un outil particulièrement adapté pour l’étude de la marche en consultation de routine. Les capteurs inertiels permettent le calcul de nombreux paramètres. L’exercice de marche de 10 m aller/retour à vitesse de confort départ arrêté donne accès aux différentes phases de la marche (initiation, croisière, demi-tour) dans des conditions de consultation de routine. Ainsi, l’objectif de ce travail est d’approcher les mécanismes d’adaptation des personnes à des perturbations à différents niveaux anatomiques des structures impliquées dans la marche. Nous abordons cette question par un phénotypage quantitatif à partir du signal de capteurs inertiels recueilli sur des patients au cours d’un exercice de marche de 10 m aller/retour en consultation clinique de routine. Nous avons étudié successivement la marche de patients atteints d’arthrose du membre inférieur comme modèle d’adaptation de la marche à la douleur, puis la marche dans la maladie de Parkinson comme modèle d’atteinte du système de la mise en place des procédures motrices, enfin, la locomotion des patients hémiparétiques à la suite d’un accident vasculaire cérébral hémisphérique comme modèle d’atteinte de la commande volontaire. Nous montrons que la douleur dans l’arthrose du membre inférieur mène à une rigidification globale de la cinématique corporelle. Cette rigidification est prépondérante sur le membre atteint. Elle traduit la perte des synergies musculaires par la mise en place de boucle-réflexe anti-douleur. Nous démontrons que ces modifications sont corrélées à la sévérité clinique de l’arthrose. Pour analyser la régularité de la marche dans la maladie de Parkinson indépendamment des variabilités inter-individuelles du patron de marche nous avons développé un outil de visualisation de l’exercice de marche. La maladie de Parkinson affecte en particulier la régularité de la marche. Notre travail apporte la preuve que cette irrégularité est corrélée à la sévérité des symptômes chez les patients atteints de la maladie de Parkinson. Nous montrons enfin qu’une lésion du cortex dans l’accident vasculaire hémisphérique provoque un changement de stratégie dans le demi-tour. Comme d’autres, nous faisons l’hypothèse que les stratégies de demi-tours sont en partie stockées dans le cortex frontal et que les hémisphères droit et gauche ne jouent pas un rôle symétrique. Nous montrons que le choix de stratégie de demi-tour est corrélé avec la survenue de chutes à 6 mois et pourrait constituer un nouvel élément pour orienter la rééducation. (...)
In the field of neurosciences, significant improvement has been made in the last decades in imaging and genotyping. However, phenotyping remains stagnant at the state of visual observation or visual grading scales. The human phenotype is made up of locomotion (gait, posture and displacement of daily living), upper-limb fine and rough movements, eye movements, language, cognition and complex social behaviors. Gait is a central function in humans, implying volitional, emotional and automatic processes. It involves the whole musculoskeletal system as well as the central and the peripheral nervous system including sensory organs. Building a gait phenotyping system implies setting up a database of gait signals of many (hundreds) of subjects and patients. This goal can be achieved with user-friendly devices deployed in routine medical practice. For instance, inertial measurement units (IMUs) are a valid tool to measure spatio-temporal gait parameters and are adapted to routine medical use. The 10-meter walking test forward and back at self-selected walking speed is adapted to routine testing at the doctor’s office. It allows for measuring gait initiation, gait cruise, gait termination and a 180° turn. In that context, beyond technical challenges, the aim of this work was to address the question How does the central nervous system adapt to an external alteration on various levels in the command chain of gait? To answer this question, we studied sequentially the IMU signal of gait spatio-temporal kinematics in lower-limb osteoarthritis as a model of gait affected by pain, in Parkinson disease as a model of a lesion of the central nervous system muscle tone regulator and finally, in post-hemispheric stroke as a model of lesions of brain structures responsible for volitional locomotion. Secondary clinical questions were How can the severity of a disease be objectively graded with gait kinematics? and How can locomotion kinematics participate in the fall risk prediction in frail populations? In osteoarthritis, we showed that pain in lower-limb osteoarthritis led to a global stiffness of the body during locomotion. This stiffness was preponderant on the affected limbs and led to the loss of muscular synergies by the establishment of anti-pain reflexes as a reaction to pain. This change was correlated with the severity of lower-limb osteoarthritis. In Parkinson disease, to analyze gait regularity independently from inter-individual gait signature, we constructed a novel gait visualization tool and show that a lesion of the muscle tone regulator in Parkinson disease affects gait regularity. This regularity was associated with disease symptoms. Finally, in stroke, we showed that a lesion in the cortex implied a change in the 180° turning stepping, a volitional task. In line with other authors, we hypothesized that locomotion patterns could be generated in the frontal cortex and that the right and left frontal cortex did not have a symmetric role. We showed specific stepping patterns associated more with risk of fall, which could constitute a new argument to orientate rehabilitation. Altogether then, this work suggests that simple measuring hardware (here IMUs), with appropriate signal processing, allowed for decomposing and quantifying complex behavioral tasks (here locomotion) in daily hospital settings

Progress in the fields of wearable sensor technologies together with specialized analysis algorithms has enabled systems for gait analysis outside labs. An example of a wearable sensor is the accelerometer embedded in a typical smartphone. The goal was to propose a system design capable of hosting existing gait analysis algorithms in a cloud environment, and tailor the design as to deliver fast results with the ambition of reaching near real-time.    The project identified a set of enabling technologies by examining existing systems for gait analysis; the technologies included cloud computing and WebSockets. The final system design is a hierarchical composition starting with a Linux VM running Node.js, which in turn connects to a database and hosts instances of the MatLab runtime. The results show the feasibility of mobile cloud based free-living gait analysis. The architectural design provides a solution to the critical problem of enabling existing algorithms to run in a cloud environment; and shows how  the graphical output of the native algorithm could be accurately reproduced in a web browser. The system can process a chunk of 1300 data points under 3 seconds for a client streaming at 128 Hz, while simultaneously streaming the real time signal.

Trabalho apresentado à Universidade Fernando Pessoa como parte dos requisitos para obtenção do grau de Licenciado em Fisioterapia
Introdução: Este estudo demonstra uma avaliação de marcha no parâmetro da velocidade média em atletas de voleibol nas idades compreendidas entre os 10 e os 16_anos. Objectivo: Contribuir para a validação do dispositivo WalkinSense® no parâmetro velocidade média da marcha humana. Resultados e Discussão: Nos critérios utilizados para o cálculo da velocidade média, o critério do ângulo de extensão do joelho mais negativo foi o parâmetro utilizado para o cálculo da Velocidade Média 3, demonstrou não ser o mais indicado, enquanto o critério do maior valor de pressão nos calcanhares utilizado para o cálculo da Velocidade Média 2, dentro das opções, o mais indicado. Conclusão: Os resultados obtidos nesta avaliação foram satisfatórios, porém moderados para a validação do dispositivo WalkinSense®. Introduction: This study demonstrates an evaluation of gait against the speed average in volleyball players between the ages of 10 and 16 years old. Objective: Contribute to the validation of the speed parameter in WalkinSense®, average human gait. Results and Discussion: The criteria used to calculate the average speed, was the criterion of the angle of knee extension most negative, that was the parameter used to calculate the Average Speed 3 proved not to be the most suitable, while the criterion of the highest pressure used in the heels to calculate Average Speed 2, within the options, as indicated. Conclusion: The results of this evaluation were satisfactory, but moderate for the validation of the WalkinSense®.

Accelerometer data can be analyzed using a variety of methods which are effective in the clinical setting. Time-series analysis is used to analyze spatiotemporal variables in various populations. More recently, investigators have focused on gait complexity and the structure of spatiotemporal variations during walking and running. This study evaluated the use of time-series analyses to determine gait parameters during running. Subjects were college-age female soccer players. Accelerometer data were collected using GPS-embedded trunk-mounted accelerometers. Customized Matlab® programs were developed that included Gaussian continuous wavelet transform (CWT) to determine spatiotemporal characteristics, detrended fluctuation analysis (DFA) to examine gait complexity and autocorrelation analyses (ACF) to assess gait regularity. Reliability was examined using repeated running efforts and intraclass correlation. Proof of concept was determined by examining differences in each variable between various running speeds. Applicability was established by examining gait before and after fatiguing activity. The results showed most variables had excellent reliability. Test-retest R2 values for these variables ranged from 0.8 to 1.0. Low reliability was seen in bilateral comparisons of gait symmetry. Increases in running speed resulted in expected changes in spatiotemporal and acceleration variables. Fatiguing exercise had minimal effects on spatiotemporal variables but resulted in noticeable declines in complexity. This investigation shows that GPS-embedded trunk-mounted accelerometers can be effectively used to assess running gait. CWT and DFA yield reliable measures of spatiotemporal characteristics of gait and gait complexity. The effects of running speed and fatigue on these variables provides proof of concepts and applicability for this analytical approach.
Master of Science
Fitness trackers have become widely accessible and easy to use. So much so that athletic teams have been using them to track activity throughout the season. Researchers are able to manipulate data generated from the fitness monitors to assess many different variables including gait. Monitoring gait may generate important information about the condition of the individual. As a person fatigues, running form is theorized to breakdown, which increases injury risk. Therefore the ability to monitor gait may be advantageous in preventing injury. The purpose of this study is to show that the methods in this study are reproducible, respond reasonably to changes in speed, and to observe the changes of gait in the presence of fatigue or on tired legs. Three analyses are used in this study. The first method called autocorrelation, overlays acceleration signals of consecutive foot strikes, and determines the similarity between them. The second method utilizes a wave transformation technique that is able to determine foot contact times. The final method attempts to determine any pattern in the running stride. This method looks for changes in the structure of the pattern. Less structure would indicate a stride that is fatigued. The results showed that the methods of gait analysis used in this study were reproducible and responded appropriately with changes in speed. Small changes in gait were observed due to the presence of fatigue. Further investigation into the use of these methods to determine changes in gait due to the presence of fatigue are warranted.

This thesis focuses on 3D marker tracking for human gait analysis. In KISS Gait Analysis System at METU, a subject'
s gait is recorded with 6 cameras while 13 reflective markers are attached at appropriate locations on his/her legs and feet. These images are processed to extract 2 dimensional (2D) coordinates of the markers in each camera. The 3 dimensional (3D) coordinates of the markers are obtained by processing the 2D coordinates of the markers with linearization and calibration algorithms. Then 3D trajectories of the markers are formed using the 3D coordinates of the markers. In this study, software which takes the 2D coordinates of markers in each camera and processes them to form the 3D trajectories of the markers is developed. Kalman Filter is used in formation of 3D trajectories. The results are found to be satisfactory.

This thesis describes the development of a tool to assist animators in doing walk cycles. In traditional animation, animators create expressive walk cycles with key poses. The process of generating walk cycles by hand is tedious and repetitive. To help animators, many researchers in computer graphics have worked on automating gait generation. However, almost all of them used methods that eliminate animator defined key poses. Although they produce realistic results, their methods are not suitable for expressive walk cycles that can be found in cartoons. The tool described in this thesis attempts to incorporate practices of traditional animators such as comparison of key poses and the use of arc into the program interface. With this tool, animators can concentrate only on setting key poses, which is the most creative task in animating expressive walk. The gait generation program can produce highly expressive walks like the double bounce walk and the sneak. With automated features of the developed tool, animators can save time and effort when animating expressive walk along a curved path.

Motion capture technology has been applied in many fields such as animation, medicine, military, etc. since it was first proposed in the 1970s. Based on the principles applied, motion capture technology is generally classified into six categories: 1) Optical; 2) Inertial; 3) Magnetic; 4) Mechanical; 5) Acoustic and 6) Markerless. Different from the other five kinds of motion capture technologies which try to track path of specific points with different equipment, markerless systems recognize human or non-human body's motion with vision-based technology which focuses on analyzing and processing the captured images for motion capture. The user doed not need to wear any equipment and is free to do any action in an extensible measurement area while a markerless motion capture system is working. Though this kind of system is considered as the preferred solution for motion capture, the difficulty for realizing an effective and high accuracy markerless system is much higher than the other technologies mentioned, which makes markerless motion capture development a popular research direction. Microsoft Kinect sensor has attracted lots of attention since the launch of its first version with its depth sensing feature which gives the sensor the ability to do motion capture without any extra devices. Recently, Microsoft released a new version of Kinect sensor with improved hardware and and targeted at the consumer market. However, to the best of our knowlege, the accuracy assessment of the sensor remains to be answered since it was released. In this thesis, we measure the depth accuracy of the newly released Kinect v2 depth sensor from different aspects and propose a trilateration method to improve the depth accuracy with multiple Kinects simultaneously. Based on the trilateration method, a low-cost, no wearable equipment requirement and easy setup human gait tracking system is realized.

Neuro-musculoskeletal impairments are a substantial burden on our health care system as a consequence of disease, injury or aging. A better understanding of how such impairments influence the skeletal system through muscle force production is needed. Clinical gait analysis lacks in a sufficient estimation of individual muscle forces. To date, joint moments and EMG measurements are used to deduce on the characteristics of muscle forces, however, known limitations restrain a satisfying analysis of muscle force production. Recent developed musculoskeletal models make it possible to estimate individual muscle forces using experimental kinematic and kinetic data as input, however, are not yet implemented into a clinical gait analysis due to a wide range of different methods and models and a lack of standardised protocols which could be easily applied by clinicians in a routine processing. This PhD thesis assessed the state of the art of mathematical modelling which enables the estimation of muscle force production during walking. This led into devising a standardised protocol which could be used to incorporate muscle force estimation into routine clinical practice. Especially the input of clinical science knowledge led to an improvement of the protocol. Static optimisation and computed muscle control, two mathematical models to estimate muscle forces, have been found to be the most suitable models for clinical purposes. OpenSim, a free available simulation tool, has been chosen as its musculoskeletal models have been already frequently used and tested. Furthermore, OpenSim provides a straight forward pipeline called SimTrack including both mathematical models. Minor and major adjustments were needed to adapt the standard pipeline for the purposes of a clinical gait analysis to be able to create a standardised protocol for gait analyses. The developed protocol was tested on ten healthy participants walking at five different walking speeds and captured by a standard motion capture system. Muscle forces were estimated and compared to surface EMG measurements regarding activation and shape as well as their dependence on walking speed. The results showed a general agreement between static optimisation, computed muscle control and the EMG excitations. Compared to the literature, these results show a good consistency between the modelling methods and surface EMG. However, some differences were shown between mathematical models and between models and EMG, especially fast walking speeds. Additionally, high estimated activation peaks and uncertainties within the estimation process point out that more research needs to be undertaken to understand the mechanisms of mathematical models and the influence of different modelling parameters better (e.g. characteristics of muscle-tendon units, uncertainties of dynamic inconsistency). In conclusion, muscle force estimation with mathematical models is not yet robust enough to be able to include the protocol into a clinical gait analysis routine. It is, however, on a good way, especially slow walking speeds showed reasonable good results. Understanding the limitations and influencing factors of these models, however, may make this possible. Further steps may be the inclusion of patients to see the influence of health conditions.