Finally, the research of classifying and identifying individuals through their walking patterns is introduced. Alive biometrical features in dynamic human gait are adopted in the intelligent shoe system. Since gait data are dynamic, non-linear, stochastic, time-varying, noisy and multi-channel, we must select a modeling framework capable of dealing with these expected complexities in the data. Using the proposed machine learning methods, support vector machine (SVM) and hidden Markov models (HMMs), we build up probabilistic models that take the information of human walking patterns into account, and compare the overall similarity among human walking patterns of several wearers. / In this thesis, we will build intelligent shoes under the framework for capturing and analyzing dynamic human gait. Existing MEMS technology makes it possible to integrate all the sensors and circuits inside a small module. In designing our intelligent shoe system, we require the following key characteristics in our system: (1) It should be convenient to wear and socially acceptable. Thus, the sensors and electronic hardware installed should not substantially change the weight and weight balance of a typical shoe, lest it alters how an individual normally walks. (2) We want to analyze a user's motion in real-time through a wireless interface to a remote laptop or other computer; we will also incorporate on-shoe data logging hardware for off-line analysis. (3) Sensors that monitor gait motion conditions may need to be attached to the insoles, in closer proximity to the foot of users. In order to investigate the problem of capturing power parasitically from normal human-body-motion for use in personal electronics applications, we also plan to develop an electromechanical generator embedded within the shoe for parasitic power collection from heel strike. / Next, we can encode specific motions to control external devices through a wireless interface. This same system architecture that allows us to classify broad categories of motion also allows the intelligent shoe to act as a programmable, low-data rate control interface. We apply the system to several successful tasks based on this platform, especially the Shoe-Mouse. By using this interface, we can operate a device with our feet. / Then, we present potential use of machine learning techniques, in particular support vector machine (SVM), and the intelligent shoe platform to detect discrete stages in the cyclic motion of dynamic human gait, and construct an identifier of five discrete events that occur in a cyclic process for precise control of functional electrical stimulation (FES). With the information of when the legs are in each phase of a gait, the timing of specific gait phase can be assessed. / Huang, Bufu. / "September 2007." / Adviser: Yangsheng Xu. / Source: Dissertation Abstracts International, Volume: 69-08, Section: B, page: 4931. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2007. / Includes bibliographical references (p. 122-131). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in English and Chinese. / School code: 1307.

Thesis (MScPhysio)--Stellenbosch University, 2015. / ENGLISH ABSTRACT: BACKGROUND: A functional gait forms an integral part of life, allowing individuals to function within their environment and participate in activities of daily living. The evaluation of gait forms an essential part of a physical examination and can help screen for physical impairments. To the researchers‟ knowledge no 3D gait analysis studies of this nature have been conducted in South Africa. South African gait analysis laboratory protocols and procedures may differ from laboratories in other countries; therefore a South African data base of normative values is required to make a valid assessment of South African children‟s gait. OBJECTIVE: The aim of this study is to describe kinematics and spatiotemporal parameters of gait of typically developed children between the ages of 6-10 years in the Cape Metropole of the Western Cape, South Africa. METHODOLOGY: A descriptive study was conducted. Twenty-eight typically developed children were conveniently sampled from aftercare facilities and schools were performed in the Cape Metropole in the Western Cape, South Africa. The three-dimensional (3D) lower limb kinematics and spatiotemporal parameters of gait were analyzed. For data capture, the lower limb Plug-in-Gait (PIG) marker placement was used. Participants were asked to walk bare footed at self-selected speed. Due to a small sample size, children were also sub-divided into two groups (Group A: 6-8 years and Group B: 9-10 years) for comparison. Means and standard deviations (SD) were calculated for all outcomes, followed by statistical tests to determine significant differences between the two sub-groups for spatiotemporal parameters and kinematics. RESULTS: There was a significant difference between the sub-groups for all the non-normalized spatiotemporal parameters. A statistical significant difference between the sub-groups for the mean hip rotation minimum values (p=0.036) was found. There was no significant difference between the sub-groups for any other kinematic parameter or when comparing the normalized spatiotemporal parameters. CONCLUSION: This study provides descriptive gait parameters that can be used for comparison or gait analysis purposes. Our results suggest that normalized spatiotemporal parameters showed no significant difference between the age groups and are consistent with international children‟s spatiotemporal parameters. Kinematic values showed significant changes with hip rotation. Older children had more external rotation at their hips. KEYWORDS: 3D gait analysis, walking, children, spatiotemporal parameters, kinematics. / AFRIKAANSE OPSOMMING: INLEIDING: „n Funksionele stap is „n essensiële deel van die lewe wat mens toelaat om in jou omgewing te funksioneer en om deel te neem aan daaglikse aktiwiteite. Evaluasie van stap is „n belangrike deel van die fisiese evaluasie en kan help om te sif vir fisiese verswakking of abnormaliteite. So ver hierdie navorsers weet, is hierdie die eerste loop analise studie van sy soort wat in Suid-Afrika onderneem is. Suid-Afrikaanse stap-evaluasie-labrotorium protokols en prosedures mag ook dalk verskil van die in ander lande. Dus is „n Suid-Afrikaanse databasis vir normale waardes van loop nodig om „n gegronde evaluasie van Suid-Afrikaanse kinders se loopgang te kan maak. DOELWIT: Die doel van hierdie studie is om die kinematika en spatiotemporale parameters van loop te omskryf in tipies ontwikkelde kinders tussen die ouderdom van 6-10 jaar in die Kaapse Metropool en om die bevindinge tussen die twee ouderdomsgroepe te vergelyk. METODE: „n Beskrywende studie is uitgevoer. Ag-en-twintig tipies ontwikkelde kinders is van skole en nasorgfasiliteite in die Kaapse Metropool in die Wes-Kaap, Suid-Afrika gewerf. Die drie-dimensionele (3D) onderste ledemaat se kinematika en spatiotemporale parameters van loop is geanaliseer. Vir data insameling is die onderste ledemaat Plug-in-Gait (PIG) merker-plasing gebruik. Deelnemers is gevra om kaalvoet teen hulle eie spoed te stap. Die kinders is in die verskeie ouderdomsgroepe verdeel, maar as gevolg van klein toetsgroepgetalle, is hulle sub-verdeel in twee groepe (Groep A: 6-8 jaar en Groep B: 9-10 jaar). Beskrywende statistiese tegnieke is gebruik vir alle uitkoms maatreëls. Gemiddeldes en standaardafwykings (SA) was bereken, om beduidende verskille tussen die ouderdomsgroepe en sub-groepe te bepaal. RESULTATE: Daar is „n beduidende verskil tussen die jonger en ouer kinders vir nie-genormaliseerde spatiotemporale parameters, asook „n beduidende verskil tussen die sub-groepe vir die gemiddelde heuprotasie minimum waardes (p=0.036). Daar was geen beduidende verskil tussen die twee groepe met die ander kinematiese parameters of met genormaliseerde spatiotemporale parameters van die sub-groepe nie. GEVOLGTREKKING: Hierdie studie verskaf beskrywende statistiese data van stap-parameters wat gebruik kan word vir vergelyking met ander kinders van dieselfde ouderdomme of loop-analise doeleindes. Ons bevindinge stel voor dat genormaliseerde spatiotemporale parameters geen beduidende bevindings aandui tussen die verskeie ouderdomsgroepe nie. Dit is ook konsekwent met internasionale kinders se spatiotemporale parameterwaardes. Kinematisie waardes het beduidende verskille in heuprotatsie getoon. Ouer kinders het meer eksterne rotasie in hulle heupe in vergelyking met jonger kinders. Soos die kinders ontwikkel, verminder die heup-anteversie en die heup beweeg vanaf interne rotasie na „n relatiewe eksterne rotasie.

Assessment of different gait patterns of daily living could provides useful information in studying one individual's stability and mobility during locomotion. As the foundation for better assessment of different gait patterns, the ability to automatically identity different patterns and walking surroundings provide valuable information for further understanding the relations between gait pattern and energy consumption. We apply Discrete Wavelet Transform (DWT) for feature generation and Fuzzy-logic based approach for designing the multi-class classifier to identify gait patterns among fiat walking, descending stairs, and ascending stairs based on continuous kinematic signals. / Falls in the aging population has always been one of the most challenging problems in public health care. We propose an automatic falling detection algorithm based on the analysis of plantar force on both feet, because plantar forces are an important parameters directly associated with postures of human locomotion. The proposed two-stage algorithm efficiently overcome the shortcomings of the widely proposed accelerometer or gyroscope based algorithms and could provide efficient assistant for automatic detection of falls once they occur. / Finally, the research of studying gait abnormalities is introduced. We develop the methodology for modeling and classifying abnormal gaits including toe-in, toe-out, over-supination, and heel walking via machine learning algorithms, hidden Markov models (HMM) and support vector machine (SVM) based on a suite of gait parameters. The trained classifiers can classify abnormal gait patterns mentioned above and the proposed methodology will make it possible to provide realtime feedback to assist persons with gait abnormalities in the development of a normal walking pattern in their daily life. / Keeping abnormal motion for long time will ultimately lead to pain in the feet, ankles, legs and skeletal disease, and badly influences the skelecton growth especially for children and adolescents. In biomedicine, gait analysis has been proved as an useful approach. in revealing helpful insights into the recognition of motion abnormalities. Analysis of gait is commonly used as a routine procedure in identifying movement or posture related abnormalities of humans and aiding the therapeutic processes. Our goal is to monitor and study gaits of humans in order that proper motion adjustments can he advised to improve their posture style and long-term well being. / Most currently utilized measurement systems for motion and gait analysis have the shortcomings of that the monitoring and analysis of motion is constrained in a limited environment and human-related assistance is essential. All of them cannot be acceptable for the purpose of long-term monitoring and studying of motion abnormalities. In this thesis, a new concept of an inexpensive, compact, and lightweight shoe-integrated platform is introduced. The shoe-integrated system is composed of a suite of sensors for wirelessly capturing gait parameters and generating well qualified analysis results. The ideal platform requires no specialized equipment or lab setup, allowing data to be collected not only in the narrow confines of a research lab, but essentially anywhere, both indoors and outdoors. / To be one of the common postural abnormalities, postural kyphosis is studied and modeled. We apply Cascade Neural Networks with Node-Decoupled Extended Kalman Filtering (CNN-NDEKF) to train the model for this binary classification problem. This proposed study is of particular significance to provide feedback in the application of postural kyphosis rectification. / Chen, Meng. / "December 2009." / Adviser: Yangsheng Xu. / Source: Dissertation Abstracts International, Volume: 72-01, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2010. / Includes bibliographical references (leaves 120-130). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.

Stair climbing is a common activity encountered in daily living. Stair ascent is a demanding task that requires a large range of motion of the joints, strong muscle strength, good cardiovascular fitness, and fine balance control. Given this, the activity can be difficult for different populations that lack muscle strength and coordination. To train and assist people in this activity, several robotic platforms have been proposed, but these limit the natural motion of the individual. For example, these devices fix the placement of the feet and reduce the natural swing of the lower limbs. This makes it difficult to manipulate the center of mass, which is crucial to stair ascent. In this dissertation, we present a novel parallel cable-driven platform in which the end effector is the user’s pelvis; the stairmill tethered pelvic assist device. This architecture allows the user to retain their natural movement and relation between the feet and the center of mass, all while applying three dimensional forces on the pelvis during continuous stair ascent on a revolving stairmill. In this work, we show the design, fabrication, and validation of this robotic system. Various force strategies were explored during stairmill ascent using this robotic platform. A characterization experiment was conducted to investigate gait performance and muscle coordination. Two simple interventions were tested to show the potential for long-term training program. This work sheds light on the different strategies of stair climbing and how we can use cable driven platforms to train and assist individuals during this challenging task. The knowledge gained by this work allows for the expansion of designing training paradigms for stair climbing with natural motion. These can assist individuals in improving their quality of life.

Multiple Sclerosis (MS) is a lifelong progressive neurologic disease of the central nervous system (CNS) that interrupts the flow of information within the brain and between the brain and the body, resulting in a variety of symptoms across the visual, sensory, motor, and autonomic functions. The concept of psychological resilience is emerging in clinical research, including research on MS, as a productive way to view the outcomes and experiences of living with a chronic disease and identify potential protective factors. The purpose of this dissertation was to examine the protective and predictive quality of psychological resilience in various domains of motor functioning. A sample of 130 patients underwent neuropsychological testing along with neurological examination at two distinct time points (baseline and 3-year follow-up). As part of each evaluation, patients were administered various tasks of motor functioning: the two-minute walk test (2MWT; a measure of gait endurance and stamina), timed 25-foot walk (T25FW; a measure of gait speed), nine-hole peg test (NHPT; a measure of upper extremity speed and coordination), grooved pegboard (G-Pegs; a measure of fine motor speed and dexterity), grip strength (Grip; a measure of upper body strength), and finger tapping test (FTT; a measure of simple motor speed), which served as this study’s outcomes. Psychological resilience, the primary predictor of interest, was operationalized as the self-reported ability of adapting well in the face of substantial adversity and significant sources of stress and was estimated using a validated self-report measure the Connor-Davidson Resilience Scale, 10 item version (CD-RISC-10). Additional predictors included mood, fatigue, demographic variables, disease variables, and magnetic resonance imaging (MRI) estimates. In contrast to our hypothesis, psychological resilience and functional outcomes were not correlated. Psychological resilience did not predict change in motor functioning over time and did not serve as a moderator between disease burden and motor functioning. As such, the present study does not provide support for psychological resilience as a protective factor for the motor system in MS or for resilience in predicting differential decline in motor functioning.

Gait analysis allows clinicians and researchers to quantitatively characterize the kinematics and kinetics of human movement. Devices that quantify gait can be either portable, such as instrumented shoes, or non-portable, such as motion capture systems and instrumented walkways. There is a tradeoff between these two classes of systems in terms of portability and accuracy. However, recent computer advances allow for the collection of meaningful data outside of the clinical setting. In this work, we present the DeepSole system combined with the different neural network models. This system is a fully capable to characterize the gait of the individuals and provide vibratory feedback to the wearer. Thanks to the flexible construction and its wireless capabilities, it can be comfortably worn by wide arrange of people, both able-bodied and people with pathologies that affect their gait. It can be used for characterization, training, and as an abstract sensor to measure human gait in real-time. Three neural network models were designed and implemented to map the sensors embedded in the DeepSole system to gait characteristics and events. The first one is a recurrent neural network that classifies the gait into the correct gait phase of the wearer. This model was validated with data from healthy young adults and children with Cerebral Palsy. Furthermore, this model was implemented in real-time to provide vibratory feedback to healthy young adults to create temporal asymmetry on the dominant side during regular walking. During the experiment, the subjects who walked had an increased stance time on both sides, but the dominant side was affected more. The second model is encoder-decoder recurrent neural network that maps the sensors into current gait cycle percentage. This model is useful to provide continuous feedback that is synchronized to the gait. This model was implemented in real-time to provide vibratory feedback to six muscle groups used during regular walking. The effects of the vibration were analyzed. It was found that depending on the feedback, the subjects changed their spatial and temporal gait parameters. The third model uses all the sensors in the instrumented footwear to identify a motor phenomenon called freezing of gait in patients with Parkinson’s Disease. This phenomenon is characterized by transient periods, usually lasting for several seconds, in which attempted ambulation is halted. The model has better performance than the state-of-the-art and does not require any pre-processing. The DeepSole system when used in conjunction with the presented models is able to characterize and provide feedback in a wide range of scenarios. The system is portable, comfortable, and can accommodate a wide range of populations who can benefit from this wearable technology.

本論文以表面肌電（Surface Electromypgraphy, SEMG）信號中動態信號能被獲取為前提，把被處理過的表面肌電信號轉變成步態參數 (gait parameters). 我們利用一些便攜式步態測量裝置，如加速度計，陀螺儀和腳踏開關和表面肌電圖測量裝置去採集步態參數。信號的處理和生物信息（身體的動態特性）轉換都加以討論和解釋，如濾波和預測肌肉的收縮等。 / 我們利用被採集步態參數作步態分析，並發現表面肌電信號內的動態信號的頻率特性能夠代表運動過程中的非恆久步態參數，如行走時的足部擺動的期間 (period of swing phase)、行走時的足部站立的期間 (period of stance phase) 和行走時的步幅期間 (period of stride)。 / 最後，我們發現可以利用線性預測 (linear prediction) 和閾值分析 (threshold analysis) 處理表面肌電信號以便獲得三種非恆久步態參數。根據我們的觀察，行走時足部擺動的期間可以被股直肌（rectus femoris, RF）的表面肌電信號捕獲，行走時的步幅期間可以被二頭肌股（bicep femoris, BF）的表面肌電信號捕獲，而行走時的足部站立的期間則可由BF和RF輸出的結果的平均值所捕獲。因此，表面肌電信號是可以作為一種獲取非恆久步態參數的工具。 / Electromyography (EMG) signal is an important quantity for describing the muscle’s activities and provides additional information in describing movement and locomotion in gait analysis. Surface electromyography (SEMG) measurement is a non-vivo technology for acquiring EMG signal. During the measurement of SEMG signals, the motion artifact is captured. Filters are applied to eliminate the frequency characteristics of motion artifact. However, this unwanted signal could be useful for obtaining the temporal gait parameters during the movement, such as the period of swing phase, the period of stance phase, and the period of stride of free walking. / In this study, accelerometers, gyroscopes and foot switches are used for the acquisition of kinematics and surface electromyography is used for measuring muscle’s activities. These measurement devices are evaluated in a gait study on lower extremity. The signal processing and conversion of bio-information (the dynamic characteristics of body) are discussed, such as filtering, and the prediction of muscle’s contraction. / Lastly, temporal gait parameters could be captured by SEMG measurement with the linear prediction process and threshold analysis. From the results, it is observed that the swing period can be captured through the SEMG measurement for rectus femoris (RF), the stride period can be captured by the SEMG measurement for bicep femoris (BF), and the stance period can be captured by the averaged result of the outputs of BF and RF. Thus, SEMG measurement could be a tool for capturing temporal gait parameters. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Chan, Chi Chong. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 67-69). / Abstracts also in Chinese. / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Literature Review --- p.1 / Chapter 1.2 --- Objectives --- p.5 / Chapter 1.3 --- Thesis Description --- p.5 / Chapter 2 --- Description for Wearable Gait Measurement --- p.7 / Chapter 2.1 --- Wearable Sensors --- p.8 / Chapter 2.2 --- Surface Electromyography (SEMG) --- p.12 / Chapter 2.3 --- Processing Unit --- p.15 / Chapter 2.4 --- Hardware Connection and Communication --- p.16 / Chapter 2.5 --- Summary --- p.20 / Chapter 3 --- Gait Analysis for Lower Extremity during Walking --- p.21 / Chapter 3.1 --- Gait Parameters Captured by Wearable Sensors --- p.21 / Chapter 3.1.1 --- Foot Switch: Walking Phase Detection --- p.22 / Chapter 3.1.2 --- Gyroscope: Frequency Response of Lower Limbs during Walking --- p.24 / Chapter 3.1.3 --- Accelerometer: Knee Joint Angle Estimation during Walking --- p.30 / Chapter 3.2 --- Analysis of Muscle Activities by SEMG signals --- p.36 / Chapter 3.3 --- Summary --- p.42 / Chapter 4 --- Temporal Gait Parameters during Walking by SEMG Measurement --- p.43 / Chapter 4.1 --- Motion Event and SEMG Signals --- p.43 / Chapter 4.2 --- Walking Phase Detection by SEMG Signals --- p.49 / Chapter 4.3 --- Temporal Gait Parameters --- p.53 / Chapter 4.4 --- Summary --- p.62 / Chapter 5 --- Conclusions, Contributions and Future Work --- p.63 / Chapter 5.1 --- Conclusions --- p.63 / Chapter 5.2 --- Contributions --- p.64 / Chapter 5.3 --- Future Work --- p.65 / Bibliography --- p.67