Search

Context: Lateral ankle sprains are one of the most common musculoskeletal injuries in the general and sporting population and as such present high-cost implications and time lost to sport and employment. Following an initial lateral ankle sprain, a high percentage of people develop chronic ankle instability with symptoms such as reduced range of motion, strength and proprioceptive deficits, episodes of giving way and instances of re-injury. Research investigating full body with multi-segmental foot kinematics and electromyography is limited thus impacting the development of successful rehabilitation and injury prevention strategies. Aim: The purpose of this research was to perform exploratory kinematic and surface electromyographic (sEMG) data analysis of the trunk, hip, knee, forefoot-tibia, forefoot-hindfoot and hindfoot-tibia between individuals with chronic ankle instability and healthy controls during walking, landing and cutting, three movements commonly associated with lateral ankle sprains Participants: Eighteen (14 males, 4 females) healthy controls (age 22.4 ± 3.6 years, height 177.8 ± 7.6 cm, mass 70.4 ± 11.9 kg) and 18 (13 males, 5 females) participants with chronic ankle instability (age 22.0 ± 2.7 years, height 176.8 ± 7.9 cm, mass 74.1 ± 9.6 kg). Participants’ data were split into the healthy control and chronic ankle instability groups based on the results of the Identification of Functional Ankle Instability questionnaire. Methods: Participants were tested during walking (Chapter 6.0), single leg landing (Chapter 7.0) and cutting (Chapter 8.0). Three-dimensional kinematics were collected using the combined Helen Hayes and Oxford Foot Model and sEMG recorded for the peroneus longus, tibialis anterior and gluteus medius. Statistical parametric mapping, discrete variable analysis and regression analysis were subsequently performed. Results: Significantly modified kinematics were observed in each of the movements performed in the chronic ankle instability group. Decreased forefoot-tibia internal rotation angular displacement was found to occur prior to initial contact in all three of the observed movements when comparing the affected limb to the healthy matched control prior to initial contact. Significantly modified electromyography was observed in the chronic ankle instability group during the cutting manoeuvre but not during the walking and landing manoeuvre. Conclusions: Key differences have been observed between groups specific to movements but also across movements. These differences are identified in not just foot and ankle kinematics but also higher up the kinetic chain in the knee, hip and trunk. Decreased forefoot-tibia internal rotation may be a variable of interest for future research due to its presence in each of the observed movements. Differences are also highlighted in the contralateral limb of the chronic ankle instability. These findings may therefore be used in the development of injury prevention and rehabilitation programmes and in the development of screening strategies. This could help to aid in the reduction in incidence of chronic ankle instability and improve the quality of life for those with chronic ankle instability.

The way surgical trainees acquire technical skills is changing in modern surgical training programmes: simulation is proposed as a key part of those changes. Arthroscopy is a surgical technique that is increasing in both incidence and technical complexity; where simulation is becoming common, but evidence is limited. Real-world performance improvements can be measured following simulation training in other fields, but equivalent measures of intra-operative performance are inadequate. Thus, although surgical simulation is popular and improves simulated performance, there is little objective evidence that it improves intra-operative performance. The original contribution of this thesis is to objectively demonstrate the transfer of simulation training into improved intra-operative technical skills. To achieve this, a systematic literature review investigated the quantitative metrics currently used to measure arthroscopic performance, identifying wireless motion analysis as a potential method to assess performance intra-operatively. Motion analysis is a recognised objective method to measure surgical activity which correlates with surgical experience, so wireless motion analysis was validated against a wired motion analysis method commonly used in simulation but not feasible for intra-operative use. Wireless motion analysis metrics were further validated with a simulated arthroscopy list: this environment allowed deliberate practice of arthroscopic sub-skills with proximate feedback for independent practice. This simulated arthroscopy list with wireless motion analysis was used in two randomised studies: the penultimate study of this thesis investigated the impact of simulated practice on the arthroscopic learning curve and showed that performance improved rapidly with independent practice but was not modified by feedback, while the final study investigated additional simulation practice during early surgical training, and objectively demonstrated that additional simulation training improved intra-operative performance compared to traditional training alone. This thesis is the first to objectively show that simulation affects intra-operative behaviour. It sets the groundwork for further investigations into efficient, cost-effective simulation and the impact of simulation training on patient outcomes.

A constantly growing corpus of interdisciplinary studies support the idea that music is a complex multimodal medium that is experienced not only by means of sounds but also through body movement. From this perspective, musical instruments can be seen as technological objects coupled with a repertoire of performance gestures. This repertoire is part of an ecological knowledge shared by musicians and listeners alike. It is part of the engine that guides musical experience and has a considerable expressive potential. This thesis explores technical and conceptual issues related to the analysis and creative use of music-related body movements in instrumental music performance. The complexity of this subject required an interdisciplinary approach, which includes the review of multiple theoretical accounts, quantitative and qualitative analysis of data collected in motion capture laboratories, the development and implementation of technologies for the interpretation and interactive use of motion data, and the creation of short musical pieces that actively employ the movement of the performers as an expressive musical feature. The theoretical framework is informed by embodied and enactive accounts of music cognition as well as by systematic studies of music-related movement and expressive music performance. The assumption that the movements of a musician are part of a shared knowledge is empirically explored through an experiment aimed at analysing the motion capture data of a violinist performing a selection of short musical excerpts. A group of subjects with no prior experience playing the violin is then asked to mime a performance following the audio excerpts recorded by the violinist. Motion data is recorded, analysed, and compared with the expert’s data. This is done both quantitatively through data analysis xii as well as qualitatively by relating the motion data to other high-level features and structures of the musical excerpts. Solutions to issues regarding capturing and storing movement data and its use in real-time scenarios are proposed. For the interactive use of motion-sensing technologies in music performance, various wearable sensors have been employed, along with different approaches for mapping control data to sound synthesis and signal processing parameters. In particular, novel approaches for the extraction of meaningful features from raw sensor data and the use of machine learning techniques for mapping movement to live electronics are described. To complete the framework, an essential element of this research project is the com- position and performance of études that explore the creative use of body movement in instrumental music from a Practice-as-Research perspective. This works as a test bed for the proposed concepts and techniques. Mapping concepts and technologies are challenged in a scenario constrained by the use of musical instruments, and different mapping ap- proaches are implemented and compared. In addition, techniques for notating movement in the score, and the impact of interactive motion sensor systems in instrumental music practice from the performer’s perspective are discussed. Finally, the chapter concluding the part of the thesis dedicated to practical implementations describes a novel method for mapping movement data to sound synthesis. This technique is based on the analysis of multimodal motion data collected from multiple subjects and its design draws from the theoretical, analytical, and practical works described throughout the dissertation. Overall, the parts and the diverse approaches that constitute this thesis work in synergy, contributing to the ongoing discourses on the study of musical gestures and the design of interactive music systems from multiple angles.

Male crickets produce a species specific song to attract females which in response move towards the sound source. This behaviour, termed phonotaxis, has been the subject of many morphological, neurophysiological and behavioural studies making it one of the most well studied examples of acoustic communication in the animal kingdom. Despite this fact, the precise leg movements during this behaviour is unknown. This is of specific interest as the cricket’s ears are located on their front legs, meaning that the perception of the sound input might change as the insect moves. This dissertation describes a methodology and an analysis that fills this knowledge gap. I developed a semi-automated tracking system for insect motion based on commercially available high-speed video cameras and freely available software. I used it to collect detailed three dimensional kinematic information from female crickets performing free walking phonotaxis towards a calling song stimulus. I marked the insect’s joints with small dots of paint and recorded the movements from underneath with a pair of cameras following the insect as it walks on the transparent floor of an arena. Tracking is done offline, utilizing a kinematic model to constrain the processing. I obtained, for the first time, the positions and angles of all joints of all legs and six additional body joints, synchronised with stance-swing transitions and the sound pattern, at a 300 Hz frame rate. I then analysed this data based on four categories: The single leg motion analysis revealed the importance of the thoraco-coxal (ThC) and body joints in the movement of the insect. Furthermore the inside middle leg’s tibio-tarsal (TiTa) joint was the centre of the rotation during turning. Certain joints appear to be the most crucial ones for the transition from straight walking to turning. The leg coordination analysis revealed the patterns followed during straight walking and turning. Furthermore, some leg combinations cannot be explained by current coordination rules. The angles relative to the active speaker revealed the deviation of the crickets as they followed a meandering course towards it. The estimation of ears’ input revealed the differences between the two sides as the insect performed phonotaxis by using a simple algorithm. In general, the results reveal both similarities and differences with other cricket studies and other insects such as cockroaches and stick insects. The work presented herein advances the current knowledge on cricket phonotactic behaviour and will be used in the further development of models of neural control of phonotaxis.

Non-contact injuries can occur when athletes use poor or inconsistent mechanics during typical sport-related movements like landing from a jump. Anterior cruciate ligament (ACL) injuries are especially devastating, and certain populations like female athletes and athletes with a previous ACL reconstruction (ACLR) are at greater risk of suffering an ACL injury, with altered biomechanical strategies being one proposed reason. Asymmetric landings where one limb experiences greater landing force can decrease joint stability and place the overloaded limb at greater risk for ACL injury. Additionally, a stiff landing, characterized by increased ground reaction force (GRF), extended joints at initial ground contact, and decreased joint flexion throughout the landing, has been proposed to increase ACL injury risk. While load distribution between limbs is a common landing assessment to determine injury risk, it is unclear what role limb stiffness plays in the likelihood of experiencing an ACL injury. Limb stiffness is simply the deformation of the limb in response to the downward force applied to the lower limb during ground contact, which can be approximated using GRF. Limb stiffness has been commonly used to assess performance in running, hopping, and jumping, however, its relationship with injury risk during landings is relatively unexplored. Past research has revealed that the ACL experiences peak strain prior to initial ground contact when the knee is at or near full extension. Additionally, expert video analyses have determined that ACL injuries most likely occur within 50 milliseconds of ground contact. It is possible that limb stiffness and limb stiffness asymmetry can be used during the early impact phase of landings to reveal ACLR- and sex-specific landing mechanics differences when the ACL appears to be most vulnerable. Moreover, game-like, sport-specific landing tasks with a greater horizontal component that load the ACL and those that divert attention away from landing strategies may uncover differences that do not appear in standard, controlled laboratory tasks. The overall goal of this project was to use limb stiffness, limb stiffness asymmetry, and related measures to analyze the early landing phase mechanics of groups at greater risk for ACL injury during game-like, sport-specific landings. First, in an ACLR cohort, greater knee power and knee work asymmetries were found when compared to healthy recreational athletes, supporting previous literature that found that athletes with an ACLR land unevenly by offloading their surgical limb. However, limb stiffness asymmetry was not different between groups, implying that the groups may have modulated limb stiffness differently between limbs. Second, minimal sex-by-task interactions were determined for landings that varied by horizontal approach prior to initial ground contact. Significant differences were found for most measures across tasks overall, however, male and female athletes displayed similar landing mechanics, indicating that expected sex-specific differences may not exist during the immediate landing phase when ACL injuries are thought to occur. Last a landing task that mimicked a ball in mid-air and diverted attention away from landing mechanics produced a sex-by-task interaction for peak impact force but no other measure. When comparing each sex-task pairing, a trend for greater peak impact force by female athletes during the distracted landing (p=0.098) was found which may indicate that future tasks with additional external focuses or another game-like component will reveal anticipated sex-specific differences. Increased time between limbs for initial ground contact for female athletes also revealed that a time-synchronized assessment of between-limb coordination may be beneficial for future research.

Background: The clinical classification of glenohumeral joint instability is characterized by presumed increased humeral translations in conjunction with symptomology. Prior research reports inconsistent trends in glenohumeral joint kinematic differences between individuals clinically classified with glenohumeral joint instability and asymptomatic controls classified with stable shoulders. Limitations surrounding clinical classification criteria and motion tracking methods likely contribute to the lack of consistent kinematic trends. Additionally, the effect of participation in repetitive, resisted, shoulder activities in individuals clinically classified with glenohumeral joint instability has not yet been examined. Purpose and approach: The purpose of this dissertation was to implement previously validated methods to clinically classify individuals with presumed glenohumeral joint instability and utilize state-of-the art kinematic assessment methods to: (1) determine the glenohumeral joint kinematic characteristics of individuals clinically classified with instability, (2) investigate the glenohumeral and scapulothoracic joint kinematic effects of exposure to repetitive, resisted, shoulder activity in this group, and (3) explore the effect of scapulothoracic rotations on humeral translations during arm raising. Results: Results from aim 1 indicated individuals clinically classified with glenohumeral joint instability possessed significantly more average anterior humeral position (0.8 mm) compared to asymptomatic matched controls during unweighted scapular plane abduction (SAB). No other kinematic differences between groups were detected. Results from aim 2 identified there was a significant decrease in average normalized contact path length (10%) between the humeral head and glenoid face during SAB and significantly less average scapular internal rotation during SAB (2.5°) and humerothoracic internal rotation (IR) (3.2°) after exposure to moderate levels of repetitive, resisted, shoulder activity. Results from aim 3 indicated there was not a significant relationship between scapulothoracic rotation and humeral translations during SAB. Summary: Findings from aim 1 of this research demonstrated that only one out of four kinematic variables used measure glenohumeral joint stability were statistically different between individuals clinically classified with glenohumeral joint instability and matched controls during unweighted SAB. These findings suggest that the magnitude of joint stability classified with passive laxity tests may not necessarily relate to dynamic joint stability. Further, perhaps more consistent kinematic differences could be identified under more vigorous task conditions. Evidence from aim 2 of this research demonstrated that participation in moderate levels of shoulder activity provoked statistically different changes in only one out of four kinematic variables used to measure glenohumeral joint stability and resulted in minimal changes (≤3.2°) in scapulothoracic kinematics during active arm raising and a simulated swimming task. These findings suggest that perhaps participation in more strenuous repetitive, resisted, shoulder activities may induce greater kinematic effects. Lastly, findings from aim 3 do not suggest the magnitude of scapular rotations affect the amount of humeral translations in individuals clinically classified with glenohumeral joint instability and imply that other factors may potentially influence glenohumeral joint stability during activity.

There are many research interests in human activity recognition, especially fall which is the major cause of serious injury for the elderly. Different technologies have been developed to detect falls, and fortunately, advancement in computer vision has attracted researchers to apply sophisticated systems for action recognition, posture estimation, and fall detection. The vision-based approach provides a non-invasive and reliable solution for fall detection among these various technologies. The overall goal of this thesis is to propose automatic human fall detection frameworks using confidence prediction and motion analysis. This thesis is composed of two pieces of work.The first work introduces a confidence-based fall detection system using multiple surveillance cameras. First, a model for predicting the confidence of fall detection on a single camera is constructed using a set of simple yet useful features. Then, the detection results from multiple cameras are fused based on their confidence levels. The proposed confidence prediction model can be easily implemented and integrated with single-camera fall detectors, and the proposed system improves the accuracy of fall detection through effective data fusion.Secondly, a flexible fall detection framework based on detecting a human object and analyzing the object’s motion is proposed. Unlike many state of the art that require predefined thresholds to detect a fall, the proposed framework localizes and tracks a person in videos via object detection and motion analysis over a time window with appropriate length. As fall events may not look the same from different view angles, a multi-view fall dataset is used to train the proposed detection method. The framework is flexible for different use cases as it could incorporate unique object detection methods and work on videos captured from different angles. The proposed framework has produced promising detection results on several other datasets that outperform two traditional methods.

Incorporating 3D understanding and spatial reasoning into (intelligent) algorithms is crucial for solving several tasks in fields such as engineering geodesy, risk assessment, and autonomous driving. Humans are capable of reasoning about 3D spatial relations even from a single 2D image. However, making the priors that we rely on explicit and integrating them into computer programs is very challenging. Operating directly on 3D input data, such as 3D point clouds, alleviates the need to lift 2D data into a 3D representation within the task-specific algorithm and hence reduces the complexity of the problem. The 3D point clouds are not only a better-suited input data representation, but they are also becoming increasingly easier to acquire. Indeed, nowadays, LiDAR sensors are even integrated into consumer devices such as mobile phones. However, these sensors often have a limited field of view, and hence multiple acquisitions are required to cover the whole area of interest. Between these acquisitions, the sensor has to be moved and pointed in a different direction. Moreover, the world that surrounds us is also dynamic and might change as well. Reasoning about the motion of both the sensor and the environment, based on point clouds acquired in two-time steps, is therfore an integral part of point cloud processing. This thesis focuses on incorporating rigidity priors into novel deep learning based approaches for dynamic 3D perception from point cloud data. Specifically, the tasks of point cloud registration, deformation analysis, and scene flow estimation are studied. At first, these tasks are incorporated into a common framework where the main difference is in the level of rigidity assumptions that are imposed on the motion of the scene or the acquisition sensor. Then, the tasks specific priors are proposed and incorporated into novel deep learning architectures. While the global rigidity can be assumed in point cloud registration, the motion patterns in deformation analysis and scene flow estimation are more complex. Therefore, the global rigidity prior has to be relaxed to local or instance- level rigidity, respectively. Rigidity priors not only add structure to the aforementioned tasks, which prevents physically implausible estimates and improves the generalization of the algorithms, but in some cases also reduce the supervision requirements. The proposed approaches were quantitatively and qualitatively evaluated on several datasets, and they yield favorable performance compared to the state-of-the-art

The ability to understand occlusion and disocclusion is critical in analyzing motion and forecasting changes. For example, when we see a car gradually blocks our view of a human figure, we know that either the car or the human is moving. We also know that the human behind the car will be visible again if we move to other positions. As many vision-based intelligent systems need to handle and react to visual data with potentially intensive motions, it is therefore beneficial to incorporate the occlusion reasoning into such systems. In this thesis, we study how we can improve the performance of vision-based deep learning models by harnessing the power of occlusion handling. We first visit the problem of optical flow estimation for motion analysis. We present a deep learning module that builds upon occlusion handling methods in classic Computer Vision literature. Our results show performance improvement in occluded regions on standard benchmarks, as well as real-world applications. We then examine the problem of view synthesis for 3D photography. We propose an inpainting method that leverages local color and depth context for novel view synthesis. We validate the proposed inpainting approach with a series of quantitative and qualitative experiments, and demonstrate promising results in predicting plausible content in occluded regions.

Over the past several decades, there have been advances in the development of complex robotic devices for daily assistance or rehabilitation. The use of such devices, however, has largely remained limited to a research setting due to the prohibitive cost and required operational engineering expertise. Likewise, dedicated biomechanics facilities perform quantitative motion analysis, contrasting the qualitative and static imaging methods which are standard in clinical care. The aim of this dissertation is to develop and validate affordable methods and devices for assessing and assisting human motion.We first present a framework for improved estimation of whole-body human kinematics with data from a single depth-camera. The algorithm incorporates biomechanical and dynamic constraints for near-real time analysis of human motion. The approach is validated against data from a ground-truth motion capture system on sit-to-stand (STS), an activity of daily living which requires significant torque generation and coordinated movement of multiple joints. We additionally present two methods for modeling the torso: a generalized relationship for the lower-lumbar angle and an optimization-based method for estimating a subject-specific model. Building on these modeling methods, we introduce a passive elastic knee orthotic device which provides bilateral knee assistance during STS. The device design and analysis integrate models of the human and device dynamics. Preliminary human subjects tests demonstrate a decrease in the human knee torque as well as positive changes in whole-body biomechanics. Finally, we introduce an affordable planar robotic manipulandum for upper limb assessment and assistance. The mechanical, electrical, and control architectures are presented, along with preliminary human subjects tests of reaching and elliptical trajectories with force field assistance under an admittance controller. A protocol for the assessment of strength and coordination is introduced and integrated with a biomechanical model of the arm. With a total material cost of less than $800, this device provides an accessible platform for clinical robotic assessment and rehabilitation.

Abstract: Motion analysis is very important and it has extensive applications in surveillance, smart rooms, and so on. Different types of sensors can be used to capture the required information. One can use eye-gaze system to record eye motion or Kinect or Leap Motion sensors to record 3d keypoints. Even a conventional digital camera is helpful for motion detection and action recognition. The spatial features of motion data change over time. These features can be the coordinates of significant points, such as fixation points of eyes, skeletal joints and so on, or the location of user defined features such as Histogram of Oriented Gradient (HOG), centre of Hough circles, centre of mass and so on. One can also analyse motion data in 1D, 2D or 3D space, depending on acquisition devices. In this thesis we studied three different applications of motion data and key-point trajectory analysis based on the different factors mentioned above. The first application is analysing eye motion to better understand team cognition between members, specifically surgeons who formed a laparoscopic surgery team. Although team cognition is believed to be the foundation for team performance, there is no direct and objective way to measure it, especially in healthcare settings. Previous studies have shown that spatial features such as overlap analysis of eye-gaze data can be a measure of team cognition. However, due to the dynamic nature of eye-gaze signals, gaze overlap calculated from spatial features is not sufficient; as team members might look at the same surgical spot at different times. Therefore, temporal feature analysis is essential . Here we studied eye-gaze signals of two surgeons throughout a simulated laparoscopic surgery task and distinguished expert teams from novice teams by the level of gaze overlap, the lag and the Recurrence Rate (RR) between two surgeons based on dual eye-tracking evidences. The results obtained in this study support the hypothesis that the top performing teams are better synchronized, show higher eye-gaze overlap and RR, and therefore demonstrate better team cognition. The second application of motion data analysis is human fall detection using a 2D video sequence. Automatic, real time fall detection techniques can improve the life quality for seniors and people with special needs, as falling down can be life threatening for these groups. Computer vision based fall detection systems require less infrastructure and is cheaper and more comfortable for users compared to smart floors or systems based on wearable devices. However, vision based systems can be less accurate and not fast enough if the set of features and detection algorithms are not selected properly or the size and generality of the training dataset does not cover different specifications. Acquiring a general training dataset is very challenging, especially for unknown surveillance regions, such as in smart houses. We proposed a robust and real time, vision based fall detection technique using only a single RGB camera. The proposed method can be applied at frame level and only uses two significant points, head and center of person. Experiment were performed on le2i fall detection dataset which is publicly available. The proposed technique can distinguish falling from everyday actions. It can also work in different indoor environments with different lighting conditions.
The last application is extracting the animation skeleton directly from 3d models regardless of its topology and initial position and orientation. This can be used to automatically animate any arbitrary 3D character, which has many applications in simulation and entertainment. Defining trajectory key-points for 3D characters without manual intervention remains a challenging problem that makes complete automation difficult. To animate an articulated 3D character, a rigging process is needed, during which an animation skeleton needs to be extracted from or be embedded into a 3D model. This tedious process is mainly done manually by expert animators. Most of the automatic rigging techniques proposed in the literature are not fully automatic nor pose invariant, i.e., a front facing model in neutral T-pose is required at the start in order to animate successfully. We proposed incorporating robust skeleton based feature detection, combined with identification of various anatomical characteristics, to extract the desired key-points along with constraint parameters needed for automatic rigging.

Surgical technical skill has a direct impact on patient health outcomes. Robotic surgical procedures present an opportunity for motion analysis-based skill assessment due to their readily available data streams detailing all manner of measurements about the tool motion. This proposes features that can help discern degrees of surgical skill. It also analyzes the importance of background contextual information in videos of surgical procedures, with regard to a crowd-sourced rater's ability to rate a surgeon's ability when background contextual information from a procedure is removed.

In this thesis research, we have explored the capabilities of a household body-tracking device - Microsoft Xbox Kinect for motion analysis and assessment to assist patient’s rehabilitation after joint replacement and repair surgeries. In particular, we have combined the motion tracking capabilities of Kinect with the 3D modelling of human body for accurate and intuitive movement assessment and feedback. Our prototype system demonstrates the effective and efficient workflow of motion capturing, data processing and analyzing and augmented reality based presentation, resulting in a design which is not only accurate, reliable and robust, but also extremely cost- and space-efficient. Furthermore, we have successfully developed a minimum viable commercial software assembling a portable motion lab that can be deployed conveniently at home, clinic office, physical therapy facilities, athlete training centers, or research institutes. We believe our system creates significant economic value for these potential customers by reducing the health care related cost, improving the rehabilitation outcome, and having the potential to prevent the excessive use of additive painkillers.

This study investigated the factors affecting the wearing of PFDs on rowing athletes’ motion and performance. Pre-experiment tests were conducted with 7 elite collegiate rowers. Both quantitative and qualitative data from the pre-test indicated that conventional Type II PFDs have significant restrictions especially in the shoulder and hip joints and were considered along with the prevailing culture of PFD nonuse by rowers. Accompanied with multi-disciplinary knowledge, a new prototype was successfully developed to minimize these restrictions, to improve mobility, comfort, and effectiveness, which was proved by post-development testing with 6 out of 7 rowers from the initial test. Possibilities for further improvements on the prototype have been proposed and preliminary flotation test was conducted to confirm the adequate buoyancy. This study implied the importance of understanding body motion when designing for specific activity. Future studies with larger sample size, field test, and metabolism measurements are expected to provide more in-depth understanding.

Introduction: Athletic performance and injury prevention are important for athletes and coaches. Different types of movement analyses have been created to aid in injury prevention and performance. The reliability of the Microsoft Kinect for movement analysis has not been widely tested. If reliable and accurate it could decrease the cost and time necessary for movement analysis. Purpose: The purpose of this study was to determine if the Microsoft Kinect is an accurate measure of knee displacement during the parallel squat when compared to the Dartfish Team Pro Software 6.0. Methods: This research used the Dartfish Team Pro Software 6.0 to validate the Microsoft Kinect as a tool to measure knee displacement. Subjects performed a parallel squat with a 7-ft long dowel rod. This exercise was used to compare value between systems. Participants were healthy members of the Athens Ohio community ages 18-30. Statistical Analysis: The intraclass correlation coefficient and paired-samples t-test were used to compare Dartfish Team Pro Software 6.0 and Microsoft Kinect. Intrarater reliability of each system was also assessed. Results: There were 29 participants in the study. The interclass correlation coefficient for Dartfish Team Pro Software 6.0 and Microsoft Kinect showed that the Microsoft Kinect had a high-reliability ICC = 0.96. Intrarater reliability for Kinect and Dartfish were .98 and .99, respectively. The mean difference between systems for measured knee displacement was 1.06 cm. The mean for the Microsoft Kinect was 49.11 ± 1.9 and 50.16 ± 96 for the Dartfish (p > 0.05). Discussion: The Microsoft Kinect is reliable against the Dartfish Team Pro Software 6.0 as a tool to measure knee displacement using the parallel squat. It appears for healthy young adults, the Microsoft Kinect is reliable for movement analysis.

This thesis considers the optimization and simulation of motions involving rigid body systems. It does so in three distinct parts, with the following topics: optimization and analysis of human high-diving motions, efficient numerical integration of rigid body dynamics with contacts, and motion optimization of a two-link robot arm using Finite-Time Lyapunov Analysis.The first part introduces the concept of eigenpostures, which we use to simulate and analyze human high-diving motions. Eigenpostures are used in two different ways: first, to reduce the complexity of the optimal control problem that we solve to obtain such motions, and second, to generate an eigenposture space to which we map existing real world motions to better analyze them. The benefits of using eigenpostures are showcased through different examples.The second part reviews an extensive list of integration algorithms used for the integration of rigid body dynamics. We analyze the accuracy and stability of the different integrators in the three-dimensional space and the rotation space SO(3). Integrators with an accuracy higher than first order perform more efficiently than integrators with first order accuracy, even in the presence of contacts.The third part uses Finite-time Lyapunov Analysis to optimize motions for a two-link robot arm. Finite-Time Lyapunov Analysis diagnoses the presence of time-scale separation in the dynamics of the optimized motion and provides the information and methodology for obtaining an accurate approximation to the optimal solution, avoiding the complications that timescale separation causes for alternative solution methods.

Ankle foot orthoses (AFOs) are commonly recommended for individuals with cerebral palsy (CP) as a means to improve gait. Goals of this dissertation were to evaluate the current efficacy of AFO use for children with CP, investigate the biomechanical mechanism of how AFOs influence gait, and describe new methods for analyzing and improving AFO outcomes as they pertain to gait. Retrospective data analysis, statistical machine learning, and simulation techniques were used to achieve these goals. Data analysis revealed that the general efficacy of AFO use was poor. However, a data driven model developed through machine learning techniques suggests that efficacy can likely be improved by using the model to recommend AFO prescriptions for individuals that are predicted to improve their gait with AFO use and refrain from prescribing AFOs for individuals whose gait will not improve with AFO use. Investigations of gait efficiency and muscle function revealed new factors that could potentially be leveraged to improve the efficacy of AFO use. Finally, an AFO design redundancy between two commonly prescribed AFOs was identified, eliminating misconceptions about the efficacy of a redundant AFO design. The techniques and conclusions presented in this dissertation have the potential to significantly improve the efficacy of AFO use for children with CP.

Abstract: Objective assessment of technical skill remains a challenging task. Paper based evaluations completed by expert assessors have been criticized for not accurately or consistently describing a surgeons’ technical proficiency due to inter-observer variability and subjective bias. In the laparoscopic or minimally invasive surgical domain, technology assisted evaluation has been shown to provide a reliable and objective measure of performance based on motion analysis, focusing on instrument movement and gestures. Aided by the miniaturization of motion tracking technology, this thesis focuses on the development of novel techniques for acquiring synchronized hand motion and eye tracking data in open surgical procedures. An overview of motor learning theory is provided as a basis for segmenting or decomposing surgical movements into constituent gestures. An empirical study investigating the learning effects of a visuospatial intensive video game as a substitute for traditional practice was performed, and showed that video gaming, can in some conditions, enhance or reinforce traditional simulator based practice. Existing motion capture techniques are reviewed along with an analysis of computational models used in high level motion analysis. A second empirical study was completed to investigate the application of one of these computer models to hand motion captured via an optical marker-less tracking device. Hidden Markov Models applied to the motion data was able to discriminate between participants emulating different levels of dexterity. Finally, the development of a technology-assisted assessment system for evaluating a surgeons’ performance based on synchronized hand motion, eye gaze and force application in open surgical techniques is presented. Several empirical studies designed to validate this iii system are described. The novel aspects of this system include the ability to capture eye gaze in a 3-dimensional environment as well as highly detailed hand motion based on a surgical glove system where 6D electromagnetic sensors are embedded. The design and assembly of this apparatus is described including an overview of the software required for achieving spatial and temporal coherence. The thesis concludes with a summary of findings and a brief discussion of planned experiments necessary to validate the clinical utility of a surgical motion and eye tracking system for both objective assessment and training purposes.

In this thesis, we have explored the use of the latest motion tracking technologies, as evident by Microsoft Xbox Kinect’s motion tracking capabilities, in combination with 3D digital human modeling and animation, multi-modality image capturing and processing, and fusion, to design a new generation of low-cost range of motion and gait analysis solutions that overcome the limitation of existing tools. The proposed solutions and our prototype systems have demonstrated accurate measurements and reliable analysis outcome compared to current clinic practices, with significantly reduced complexity and cost. Furthermore, it eliminates the need for expensive effort for pre- and post- processing of data, and also the need for a large lab space for placing the camera array. As a result, it is particularly suitable for deployment directly at doctor/therapist offices. Without having to send patient to the motion labs, which could be far away and expensive in most places, it gives them a tool to quickly and conveniently capture and access the range of motion and gait result for enhanced diagnosis, treatment, and rehabilitation for joint problems, e.g., for joint replacement and repair patients.Specifically, we have developed a touch-free solution for measuring the joint range of motion in the human body to address the clinic needs of evidence to support accurate diagnosis, treatment and rehabilitation for joint problems, and improvediiiinteraction and among patient, doctor, therapist and etc. In our approach, we will gather motion data captured through the motion tracking device Kinect and real- timely process motion data to obtain range of motion in enhanced accuracy and reproducibility. We have completed a prototype, which illustrates our entire work flow. Our preliminary experiment results have shown our system provides reliable and effective analysis of range of motions.We have also started the design and development of a Kinect-based gait analysis system to provide a broad range of high quality body motion analysis without having to depend on high cost equipment. We have completed all function components of the prototype and have use it to demonstrate the correctness, effectiveness and reliability of such an ultra-low cost solution in practice. Our preliminary experiments for feasibility study have shown consistent reproducibility and accuracy.In Summary, we have developed prototypes for range of motion and gait analysis, and have carried out experimental study to show their capabilities. We believe these new techniques will improve the current practice in clinical body motion analysis.

Medical images are being increasingly used and facilitate various applications such as computer aided diagnosis, quantitative functional analysis, treatment planning and image-guided interventions. In those applications, multiple images are usually acquired from multiple/single subjects at different time points or in real-time for comparison, progression monitoring or guidance and management during interventions. Such acquisition offers a unique opportunity to study patient anatomy in the context of multiple realizations, and permits both the construction of patient (or sub-population cohort) specific context and motion analysis. To fully benefit from those applications, shape and motion analysis are needed. The former can be considered in the general research area of shape estimation and segmentation from an image processing perspective; the latter addresses motion estimation and prediction, two problems of ultimate clinical importance. These are challenging questions, especially when images are acquired with different modalities, subject to low SNR and/or intensity and contrast variations as a consequence of contrast dynamics. It is also a well-admitted challenge to predict anatomical motion due to the complexity of motion pattern combined with the curse of dimensionality from high-dimensional anatomy representations. To this end, this thesis aims to address those challenges. One key and fundamental component of our development is the extraction of robust shape features from images, based on the observation that boundaries/shapes are usually better preserved and more consistent even under changing or challenging imaging conditions. The main contribution of this thesis is by developing variational frameworks to extract shape features automatically and reliably and building proper low-dimensional embeddings, based on which motion estimation and prediction methods are developed. The first part of this thesis focuses on solving motion estimation/tracking based on extracting and registering shape features. We have developed and validated a robust motion estimation method, by registering shape features extracted from a variational segmentation method. Extracted with length and temporal shape regularizations, the shape features are robust to intensity and contrast variations and low SNR. The continuous representation of the shape features further eliminates the needs and risks of large errors of building and registering explicit correspondences among features. To evaluate its clinical value, we have applied the proposed method to compensate respiratory motion in MR urography, and quantitatively evaluated its performance on estimating functional renal parameters. To further complement the shape extraction step in the proposed method, while also serving as a fundamental methodology development, we have further developed a unified segmentation framework by incorporating a novel sparse composite shape prior, which is especially advantageous for shape extraction when images are subject to high noise and/or with signal voids. We have evaluated and shown clinical values of this framework in solving various challenging segmentation problems including corpus callosum segmentation in 2D MR, liver segmentation in 3D CT and left ventricle segmentation in Cine MRI. We then address the shape extraction problem when point clouds are acquired using photogrammetry systems in image-guided radiotherapy. We represent and reconstruct continuous shapes/surfaces from acquired point clouds by minimizing a regularized variational functional, such that the resulting surfaces are robust to noise and missing measurements. To further speed-up the reconstruction, we have developed a real-time surface reconstruction method based upon the overcomplete nature of the respiratory motion, by representing each point cloud as a sparse combination of training set and propagate such linear relation to the continuous surface space. Finally, we investigate how to efficiently model temporal dynamics of high-dimensional motion by learning their low-dimensional embeddings. We have developed a unified prediction framework on high-dimensional states by applying manifold learning to construct a low-dimensional feature sub-manifold, where efficient prediction can be performed. A pre-image estimation method is also explored to map the predicted value in the sub-manifold to its original high-dimensional space. Our developed methodologies for shape and motion analysis are general and their clinical values are beyond the DCE-MRI and image-guided radiotherapy as demonstrated in this thesis. We also expect the development of robust shape descriptors to have substantial impact in the general field of computer vision when motion estimation and prediction are needed, especially when images captured are under non-ideal conditions.

The etiology of nonspecific low back pain is sparsely understood. To better understand the contributing factors to nonspecific low back pain, there are often common concurrent pathologies that are investigated to determine their functional relationship to low back pain. One such pathology, investigated further in this thesis, is tight hamstrings. Specifically, the effect of hamstring length on pelvic position during gait and activities of daily living under normal and altered spinal position were investigated as part of this study in a motion analysis lab. First, a marker validation study was conducted to ensure the accuracy of sagittal spinal measures of lumbar lordosis, thoracic kyphosis, and sagittal vertical axis, which are calculated using skin markers. Lateral x-rays taken by the EOS bi-planar scanner were used to measure both clinical and marker measures. Sagittal spinal measures were also output by the built-in sterEOS software. These measures were compared and found to be accurate within clinical requirements, despite inaccuracy of individual marker placement in identifying intended spinal anatomy. After validating the accuracy of spinal measures of interest for this study, kinematics of the pelvis and spine were analyzed during normal gait under two conditions: at a normal and altered spinal position. This revealed a unique pelvic compensation pattern in those with tight hamstrings to changes in lumbar lordosis. While other study participants exhibited varied pelvic responses to changes in lumbar lordosis, those with tight hamstrings responded with a -0.7° ± 1.6° decrease in pelvic tilt for every 1° of decreased lumbar lordosis (R2 = 0.94). Finally, a similar kinematic analysis was conducted during stair ascent and descent. The results of this analysis, however, revealed a more random pelvic compensation pattern to changes in lumbar lordosis even among those with tight hamstrings. A kinematic and kinetic analysis of the angles, moments, and powers at the hip, knee, and ankle during stair ascent and descent also revealed no significant differences between those with and without tight hamstrings, with the exception of hip kinematics during swing phase of stair ascent (p = 0.047).

Articular cartilage defects in the knee are common, particularly among athletes. Because of the limited healing capacity of cartilage, defects are at risk for progression towards osteoarthritis. Many treatment options exist, however, choosing the optimal treatment plan for each patient remains a challenge because the natural history of articular cartilage defects is not fully understood.The purpose of this dissertation was to investigate measurable biomechanical factors that may be associated with cartilage degeneration in the knee as an early step toward improving patient care and preventing progression. First, we determined the sensitivity of joint kinematics to various calculation techniques in order to assist with selection of appropriate data analysis protocols. Next, we explored joint moment symmetry in healthy adult populations during walking, which revealed a large magnitude and prevalence of asymmetry and provided a context for interpreting the results of additional investigations.We then evaluated joint moments and intersegmental joint reaction forces in the knees of collegiate football linemen during sport-specific movement patterns to identify a possible contributor to the increased risk of cartilage injuries and degenerative joint diseases in football linemen. The magnitude of forces and moments produced during lineman-specific activities did not exceed those during walking or jogging; however, the deeper knee flexion angles at which they occurred may place unconditioned cartilage on the posterior femoral condyles at risk.We developed simplified finite element models of tibiofemoral contact to investigate the effects of bone geometry, meniscal deficiency, joint loading, defect size and their interactions on stress and subchondral bone contact, which have been suggested as primary means of defect progression. We identified different sensitivities to bone geometry in medial and lateral compartment models. Significant interaction effects suggest the benefit of considering multiple parameters when assessing the risk of defect progression.Finally, we used motion capture and magnetic resonance imaging to determine the effects of a season of football play, and subject-specific movements, on cartilage health in the knees of asymptomatic linemen. We identified cartilage abnormalities in 60% of our subjects at preseason and observed decreases in cartilage health in more than half of our subjects over the season. More years of football experience was a significant predictor of lower glycosaminoglycan (GAG) concentration at baseline. Greater vertical joint reaction forces during walking and greater frontal plane moments during lineman-specific movements were associated with higher preseason GAG concentration, but with a greater decrease in concentration over the season. These results provide additional evidence that cartilage may adapt to regular loading; however, the loading that occurs during a season of football may disrupt cartilage homeostasis and contribute to cartilage degeneration.This dissertation advances our understanding of factors which may influence cartilage defect progression and has identified potential predictive relationships between subject-specific parameters and changes in cartilage health. The research presented here lays the groundwork for future investigations that will continue to expand our understanding of defect progression with the goal of assisting with surgical decision-making and providing patients with an optimal treatment plan to restore joint function and prevent degeneration.

The study investigated how the subjects, 18 children with spastic diplegia aged 7-14 years, attained sit-to-stand (STS). The children were divided into two groups and three STS conditions: 1) those who could attain STS independently (I-STS), 2) those who could not attain STS independently (D-STS), and 3) subjects from the D-STS condition who could successfully attain STS with the walker (W-STS). The results showed that I-STS had more mean maximum horizontal location of the upper body and knee than the hip. All body segments of D-STS followed the same model as the I-STS condition, but they moved with less magnitude than I-STS. W-STS presented both pattern and magnitudes relatively similar to I-STS. Furthermore, I-STS showed the highest mean maximum horizontal and vertical velocities of body segments, when compared with the other STS conditions. W-STS performed the mean maximum horizontal and vertical linear velocities of all selected segments close to D-STS did.

Skills assessment in Minimally Invasive Surgery (MIS) has been a challenge for training centers for a long time. The emerging maturity of camera-based systems has the potential to transform problems into solutions in many different areas, including MIS. The current evaluation techniques for assessing the performance of surgeons and trainees are direct observation, global assessments, and checklists. These techniques are mostly subjective and can, therefore, involve a margin of bias. The current automated approaches are all implemented using mechanical or electromagnetic sensors, which suffer limitations and influence the surgeon’s motion. Thus, evaluating the skills of the MIS surgeons and trainees objectively has become an increasing concern. In this work, we integrate and coordinate multiple camera sensors to assess the performance of MIS trainees and surgeons. This study aims at developing an objective data-driven assessment that takes advantage of multiple coordinated sensors. The technical framework for the study is a synchronized network of sensors that captures large sets of measures from the training environment. The measures are then, processed to produce a reliable set of individual and composed metrics, coordinated in time, that suggest patterns of skill development. The sensors are non-invasive, real-time, and coordinated over many cues such as, eye movement, external shots of body and instruments, and internal shots of the operative field. The platform is validated by a case study of 17 subjects and 70 sessions. The results show that the platform output is highly accurate and reliable in detecting patterns of skills development and predicting the skill level of the trainees.

Human action recognition is used to automatically detect and recognize actions per- formed by humans in a video. Applications include visual surveillance, human-computer interaction, and robot intelligence, to name a few. An example of a surveillance application is a system that monitors a large public area, such as an airport, for suspicious activity. In human-machine interaction, computers may be controlled by simple human actions. For example, the motion of an arm may instruct the computer to rotate a 3-D model that is being displayed. Human action recognition is also an important capability of intelligent robots that interact with humans.General approaches to human action recognition fall under two categories: those that are based on tracking and those that do not use tracking. Approaches that do not use tracking often cannot recognize complex motions where movement of different parts of the body is important. Tracking-based approaches that use motion of different parts of the body are generally more powerful but are computationally more expensive, making them inappropriate for applications that require real-time responses.We propose a new approach to human action recognition that is able to learn various human actions and later recognize them in an efficient manner. In this approach, motion trajectories are formed by tracking one or more key points on the human body. In particular, points on the hands and feet are tracked. A curve is fitted to each motion trajectory to smooth noise and to form a continuous and differentiable curve. A motion curve is then segmented into “basic motion” segments by detecting peak curvature points. To recognize an observed basic motion, a vector of curve features describing the motion is created, the vector is projected to the eigenspace created during PCA training, and the action most similar to a learned action is identified using the k-nearest neighbor decision rule.The proposed approach simplifies action recognition by requiring that only a small number of points on a subject's body be tracked. It is shown that the motion curves obtained by tracking a small number of points are sufficient to recognize various human actions with a high degree of accuracy.Furthermore, the proposed approach can improve the recognition power of other ap- proaches by recognizing detailed basic motions, such as foot steps, while introducing ef- ficient tracking and recognition compared to previous approaches. Recognition of basic motions allows a high-level recognizer to recognize more complex or composite actions by using the proposed system as a low-level recognizer.Contributions of this work include reducing each video frame to a few key points on the subject's body, using curve fitting to smooth trajectory data and provide reliable seg- mentation of the motion, and efficient recognition of basic motions using PCA.

Although patellofemoral pain syndrome (PFPS) is known as one of the most common injuries in a physically active population, its influence on movement coordination and neuromuscular functions has not been fully understood, and previous studies have only reported discrete kinematics of a single joint. The aim of this study was to perform simultaneous investigation of movement coordination between two joints during stair ambulation combined with muscular activation patterns and kinetic characteristics in those with and without PFPS. Our results revealed that movement coordination patterns are different in the frontal and transverse planes in PFPS individuals, with more restricted movement variability, indicating the reduced ability to utilize multiple strategies for performing stair ambulation tasks. Furthermore, these movement coordination alterations were present along with the increased knee abduction moment and impulse, and altered muscle activation patterns of the lower extremity muscles. Our results may provide additional support for implementing rehabilitative exercise programs that promote a wider range of movement strategies while reestablishing proper balance of the lower extremity muscles activation patterns.

The purpose of this dissertation was to analyze shoulder joint kinematics and electromyographic activities of wheelchair propulsion between two stroke patterns. Twenty physical therapy students (14 females and 6 males, age 27.4 ± 5.9 years, body mass 64.41 ± 9.37 Kg and body height 169.32 ± 9.12 cm) participated. Eleven reflective markers were placed on thorax and right scapula, humerus, third metacarpophalangeal joint and wheelchair axle. Surface electrodes were placed on right pectoralis major, anterior and posterior deltoids, infraspinatus, middle trapezius, biceps brachialis long head and triceps brachialis. Participants propelled a standard wheelchair on a stationary roller system at 0.9 m/s and 1.8 m/s with semicircular (SC) and single loop (SL) stroke patterns for 20 seconds. Three-dimensional body movement and muscle activities were recorded at 100 and 1000 Hz, respectively. All data were compared for differences between two patterns and two speeds using 2-way repeated measures ANOVA (α < .05). Results showed longer drive phase and shorter recovery phase in SC when compared to SL, with no difference found on cycle time. Smaller release angles in SC caused longer angle ranges of hand contact on the pushrim while initial contact angles did not change. During drive phase, smaller scapular protraction range of motion (ROM) was found in SC. Shoulder abduction in drive phase was larger in terms of the maximal angle and ROM. In the recovery phase, minimal scapular tilting, protraction, and shoulder abduction and internal rotation were larger in SC when compared to SL pattern. Shoulder linear velocities and accelerations were higher in both phases for abduction/adduction and flexion/extension in SC. For SC pattern, pectorals major and middle trapezius showed lower activities during drive phase while posterior deltoid and triceps showed higher activities during both phases when compared to SL. Although posterior deltoid and triceps muscles work harder in SC pattern, longer drive phase and lower muscle activities in pectorals major and middle trapezius during the drive phase may make SC the better stroke pattern in wheelchair propulsion when compared to SL.