Proper Gait Research Articles

Gait ataxia in alcohol use disorder: A systematic review.

A severe and long-term alcohol use can have adverse effects on lower limb function. Over time, some individuals may develop gait ataxia, which refers to the impairment of controlled lower body movements that are important for walking and maintaining proper gait. Gait ataxia is well-documented in patients who have been diagnosed with alcohol-related Wernicke-Korsakoff syndrome (WKS); however, less is known on how common ataxia is among patients with alcohol use disorder (AUD) without WKS. To date, no study has systematically reviewed the evidence focusing on patients suffering only from AUD. Our aim was to perform a qualitative synthesis of the existing literature examining behavioral signs of gait ataxia among abstinent patients with AUD. Two facets were created encompassing keywords for "alcohol use disorder" and "measures of gait ataxia." Databases, including EMBASE, APA PsycInfo, Medline, and Cochrane Library, were searched for studies, and a quality assessment was performed. Ten studies were identified (37 ≥ ns ≤ 247), which were all rated as being of moderate (N = 7) to good quality (N = 3). The age range was 31.4-53.4 years (weighted mean age: 53.6 years), and 78.3% of the participants were male. Eight studies found that patients with AUD and without WKS exhibited behavioral signs of gait ataxia. Although there is evidence of gait ataxia among patients with AUD, heterogeneous results and methodological shortcomings such as lack of screening for neurocognitive deficits deem these findings preliminary and highlight the need for more research in the future. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

What Is the Proper Gait Pattern for People With Paraplegia Who Wear Powered Exoskeletons?: Re-examining gait patterns of existing powered exoskeletons

Unlike able-bodied humans, people with paraplegia are unable to voluntarily move their legs or waist, so they cannot control their center of gravity (CoG). Powered exoskeletons have been regarded as one of the most significant lower-limb rehabilitation devices in recent years, and existing powered exoskeletons have already been used to assist people with paraplegia to walk similarly to a normal human gait. However, it may not necessarily be the best option for people with paraplegia to walk like able-bodied people because powered exoskeletons are much more limited in their degrees of freedom (DoF) than an able-bodied human body and cannot help people with paraplegia voluntarily control their CoG. Therefore, there is a need for research specifically targeting people with paraplegia wearing powered exoskeletons to develop their own appropriate gait pattern method according to the characteristics of powered exoskeletons. In this study, a new gait pattern, called <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">forward-leaning walking</i> ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">FLW</i> ), is proposed for the motion control of powered exoskeletons. The proposed method enables transfer of the CoG to the leading leg by using the height difference between the leading and trailing legs. The effectiveness of the proposed gait pattern method was evaluated by tilting the user’s trunk by 1. The initial frontal acceleration of the user’s CoG increased by 12–16% each time the angle of the tilt of the trunk increased by 1 during the preswing phase. In addition, several sensors were used for quantitative evaluation, and the successful weight shift of the legs of people with paraplegia was monitored. The metabolic cost was also measured to verify the efficiency of the proposed gait method as people with paraplegia walked with powered exoskeletons, the oxygen consumption cost was more than halved, and the locomotion speed tripled compared to the first training session.

Design Optimization of the Lift Mechanism in the Robotic Walking Training Device Using the Engineering Design Methodology

Partial paralysis caused by spinal cord injury (SCI) or stroke are two of the most prevalent forms of physical disability. Through proper gait training, people with incomplete SCI have more potential to retain or regain the ability to walk than those with complete SCI. To help patients who have these disabilities regain the function of walking unassisted, the robotic walking training device (RWTD) has been developed to perform gait rehabilitation. This research plays a pivotal role in advancing medical robotic technology and gait rehabilitation by conducting a comprehensive evaluation and comparison of three lift mechanisms. Specifically, the lift mechanisms are designed to reposition a patient, using the RWTD, from a supine to a vertical position. Addressing a crucial gap in supporting and placing patients in gait rehabilitation devices, design optimization was performed using the engineering design process. This approach utilizes sophisticated techniques, including CAD modeling, motion analysis, structural analysis using finite element analysis, and a Pugh decision matrix. The findings offer valuable insights for optimizing lift mechanisms for the RWTD, contributing to the enhancement of patient-centric care. This research ensures a focus on safety, efficiency, and comfort in the gait rehabilitation process, with broader implications for the evolution of medical robotic devices.

Machine Learning for Human Motion Intention Detection.

The gait pattern of exoskeleton control conflicting with the human operator's (the pilot) intention may cause awkward maneuvering or even injury. Therefore, it has been the focus of many studies to help decide the proper gait operation. However, the timing for the recognization plays a crucial role in the operation. The delayed detection of the pilot's intent can be equally undesirable to the exoskeleton operation. Instead of recognizing the motion, this study examines the possibility of identifying the transition between gaits to achieve in-time detection. This study used the data from IMU sensors for future mobile applications. Furthermore, we tested using two machine learning networks: a linearfFeedforward neural network and a long short-term memory network. The gait data are from five subjects for training and testing. The study results show that: 1. The network can successfully separate the transition period from the motion periods. 2. The detection of gait change from walking to sitting can be as fast as 0.17 s, which is adequate for future control applications. However, detecting the transition from standing to walking can take as long as 1.2 s. 3. This study also find that the network trained for one person can also detect movement changes for different persons without deteriorating the performance.

Neurorehabilitation Method for Preventing the Collapse of Internal Model: Verification of Unconscious Motor Change Caused by Implicit Error Involved in Multimodal Sensory FBs(Feedbacks).

It remains challenging for patients with stroke to regain the proper gait they had before stroke onset. In this study, one of the significant reasons for the problem is assumed to be the collapse of patients' internal model. Therefore, this study aimed to propose a method for lower-limb rehabilitation in the acute phase to prevent the collapse of the internal model and simultaneously promote voluntariness.To verify the effect of the proposed method and determine the optimal FB conditions, a supine pseudo-walking system has been developed providing multimodal sensory FBs, including somatosensory FB (walking on the walking robot), load transfer FB (plantar vibration provided by the robot), and visual FB (walking scene shown in virtual reality).First, the implicit error range of each sensory FB that was unable to recognize was investigated according to Weber's law. Subsequently, 12 participants were asked to perform 12 steps of walking exercise in the system to evaluate the unconscious motor-changing effect due to different/multimodal FBs. Errors were presented at the 7th step in either three FBs in the ratio determined or their combinations (two or three types of FBs simultaneously changing).Second, to examine whether the effect can last and influence the motion after the training, a pre-post comparison test was performed with the training in the middle, wherein implicit errors were presented in the three kinds of FBs.The results show that presenting implicit error can cause motor change, and multiple types of FBs may have a better effect. The statistical data of 12 participants of the pre-post comparison test show that the average integrated electromyogram in the post-test appears significantly larger than that in the pre-test. The result indicates that the motor-changing effect of the proposed method has a subsequent influence on motion and shows the possibility of the future application of the proposed method to the rehabilitation field.

Upper limb movement during locomotion in patients with neurological disorders: a wearable-based approach

The contribution of upper limbs during walking is crucial in maintaining balance and stability: humans, in fact, swing each upper limb in phase with contra-lateral lower limb at their comfortable walking speed, with a ratio of 1:1 [1], thus counteracting the angular momentum, reducing the energy expenditure, and improving the overall gait efficiency [2,3]. In people who suffer from neurological disorders, like Multiple Sclerosis (MS) or Traumatic Brain Injury (TBI), upper limb function is often impaired, resulting in differences in arm swing frequency, amplitude, symmetry, as well as in the ratio between step and arm swing frequency, compared to healthy people [2]. Although it is known that a neurological rehabilitation program should focus on the recovery of a proper gait also emphasizing the contribution of upper limbs [1], few studies analyzed the contribution of the upper arms during locomotion in people with neurological disorders. The aim of this study is to explore a wearable sensor-based protocol for the assessment of the upper limb movement in patients with MS and TBI compared to healthy participants. Thirty-three patients suffering from MS (24 F; 49.8±9.3 y; EDSS 4.0±1.6), 8 patients suffering from TBI (2 F; 32.3±14.1 y; DGI 18.3±5.9), and 20 healthy adults (HC) (9 F; 37.0±19.9 y) were enrolled in this study (CE/PROG.700). Each participant performed a 10-meter Walking Test (10mWT) while wearing 5 synchronized inertial measurement units (IMUs) located at sternum, both wrists and distal tibiae level. Stride segmentation was performed using tibiae-mounted IMUs and the following parameters were estimated from wrist- and sternum-mounted IMUs [4]: arm oscillation frequency, peak and range of the angular velocity norm, as well as step to arm oscillation frequency ratio for the right (RW) and left (LW) wrists (freq, gyro peak, gyro range, Step-freq ratio). The range of the sternum cranio-caudal angular velocity was also obtained (sternum gyro range), together with a measure of symmetry (Symmetry Index [5]) between RW and LW angular velocity ranges. After checking for data normality, a one-way ANOVA was used to test for significant differences among MS, TBI, and HC. Post-hoc analysis was performed using Bonferroni-Holmes correction for multiple comparisons. Results are depicted in Fig. 1 (the larger the spider, the better the performance). Significant differences were found in frequency, symmetry, and angular velocity parameters among the MS, TBI, and HC. Patients with TBI seem to be characterized by a reduced performance with respect to MS, except for the step to arm swing ratio. The proposed assessment protocol demonstrates a potential clinical value in discriminating not only patients with neurological disorders from HC, but also patients with MS from TBI, who showed significant differences in terms of swing frequency, angular velocity range, and symmetry. This information could be useful in tailoring rehabilitation treatments involving the upper limbs during gait exercises [1]. Although in the present study, no significant difference was found among the three populations walking speed, the effect of the latter on the proposed metrics should be better investigated in further studies.

HAPTR2: Improved Haptic Transformer for legged robots’ terrain classification

The haptic terrain classification is an essential component of a mobile walking robot control system, ensuring proper gait adaptation to the changing environmental conditions. In practice, such components are a part of an autonomous system and thus have to be lightweight, provide fast inference time, and guarantee robustness to minor changes in recorded sensory data. We propose transformer-based HAPTR and HAPTR2 terrain classification methods that use force and torque measurements from feet to meet these requirements. For reliable comparison of the proposed solutions, we adapt two classical machine learning algorithms (DTW-KNN and ROCKET), one temporal convolution network (TCN), and use the state-of-the-art CNN-RNN. The experiments are performed on publicly available PUTany and QCAT datasets. We show that the proposed HAPTR and HAPTR2 methods achieve accuracy on par or better than state-of-the-art approaches with a lower number of parameters, faster inference time, and improved robustness to input signal distortions. These features make HAPTR and HAPTR2 excel in terrain recognition tasks when considering real-world requirements.

Modern Aspects of Endoprosthetics in Hip Joint Osteoarthritis

The surgical procedure for an artificial joint – the endoprosthesis implantation significantly improves the patient’s quality of life and has a significant impact on his rehabilitation optimizing, necessary for a normal life in the future. Due to its anatomical characteristics and biomechanical ratio, the hip joint plays a critical role in maintaining body balance and proper gait. Endoprosthesis is used in a significant hip joint function disorder. Indications for the endoprosthesis may be degenerative changes, various injuries or some congenital disorders. A degenerative change, like coxarthrosis, leads to surgery due to its progression in almost all cases. Depending on an operator and a patient, this procedure can be performed in several ways depending on the chosen access to a thigh. Today, the direct access is most commonly used from the front, side, and rear. The difference is in a place of entry into the joint and the damage to the muscles caused by the procedure itself. Physiotherapy is required after the procedure. The process of physiotherapy begins immediately after the surgery, on the first day after the procedure. It is believed that, in the period of three to four months after rehabilitation, the patient can return to almost all the previous usual activities, of course, after examination and consultation with the doctor. Aim. To analyze the literature data on modern aspects of arthroplasty in the hip joint osteoarthrosis Results. Various arthroplasty complications in coxarthrosis can be associated with errors in the surgery techniques, as well as the lack of patients’ motivation. The number of studies concerning quality of life and therefore health is growing, and the research results can be used as relevant ones to represent the state of each individual group.

ASSESSMENT OF FOOT POSTURE CHANGES

Abstract. The structural complexity of the foot emerges from its varied and precise characteristics. This small structure can adapt to different situations, walking on uneven terrain such as mountainous areas or running on flat ground while keeping the body in balance on small surfaces. For humans, the foot is the basis of support and propulsion during walking and is also involved in providing the support and flexibility needed to achieve efficient weight transfer. Therefore, the correct distribution of loads and pressures on the surface of the foot is crucial to ensure lower limb function and proper gait and balance under static and dynamic conditions. A major limitation of studies investigating the relationship between foot posture and lower-limb muscle function is that there is currently no universally accepted method of classifying foot posture. There is a need for a dynamic way of classifying foot posture, which can be achieved by providing a reliable assessment process for foot posture changes. In clinical practice, the assessment of static foot posture is a common approach to classifying the type of foot to identify possible change-related aetiological factors and recommend therapeutic intervention. This approach involves examining the structural alignment or posture of the foot and the characteristics of its dynamic function to theoretically establish the mechanisms predisposing to pathological changes. In this context, this paper highlights the need for a reliable assessment of changes in foot posture, so our purpose is to present modern assessment and classification methods commonly used in clinical practice and research.

Assessment of foot posture changes

The structural complexity of the foot emerges from its varied and precise characteristics. This small structure can adapt to different situations, walking on uneven terrain such as mountainous areas or running on flat ground while keeping the body in balance on small surfaces. For humans, the foot is the basis of support and propulsion during walking and is also involved in providing the support and flexibility needed to achieve efficient weight transfer. Therefore, the correct distribution of loads and pressures on the surface of the foot is crucial to ensure lower limb function and proper gait and balance under static and dynamic conditions. A major limitation of studies investigating the relationship between foot posture and lower-limb muscle function is that there is currently no universally accepted method of classifying foot posture. There is a need for a dynamic way of classifying foot posture, which can be achieved by providing a reliable assessment process for foot posture changes. In clinical practice, the assessment of static foot posture is a common approach to classifying the type of foot to identify possible change-related aetiological factors and recommend therapeutic intervention. This approach involves examining the structural alignment or posture of the foot and the characteristics of its dynamic function to theoretically establish the mechanisms predisposing to pathological changes. In this context, this paper highlights the need for a reliable assessment of changes in foot posture, so our purpose is to present modern assessment and classification methods commonly used in clinical practice and research.

Modeling and Simulation of an Assistive Exoskeleton Leg System With Knee Osteoarthritis and Quadriceps Weakness

Abstract The most common human locomotion problems such as quadriceps weakness, knee osteoarthritis can be healed up by using exoskeleton mechanisms with proper control systems. However, these kinds of abnormalities cannot be easily modeled in terms of engineering perspectives due to a lack of adequate data or unknown dynamics. Also, nature always seeks minimum energy as well as biology, which means that the unknown dynamics can be built by using this phenomenon. In this study, a new system dynamic model had been involved in designing a simple single-legged exoskeleton robot mechanism and its control system to assist partially disabled individuals to improve their quality of locomotion. To determine the specific features of the human gait disorders to interpret their nature in the computer-aided simulation environment, knee osteoarthritis, and quadriceps weakness, which are the common types of such problems, have been chosen as the main interests for this study. By using the lower limb model with anthropometric data, the simulations of disorders have been realized on matlabsimscape environment, which enables us to model the entire exoskeleton system with the three-dimensional parts of the human body. A model of a leg with the disorder was able to be obtained with the utilization of feedback linearization, which is one of the examples of minimum principles in the control theory. A proper gait cycle is achieved with the exoskeleton application and separately for the leg, with approximately 10 deg deviation from the natural property in knee flexion. Finally, it can be seen that the system conversion into such problematic cases with or without an exoskeleton system is accomplished.

Dynamic Lower-Limb Alignment Focusing on Gait Stability

Introduction: Prosthetic limbs must be developed with proper alignment to facilitate safe and efficient gait patterns. This study aimed to identify factors impacting clinically proper gait patterns by objectively evaluating them in persons with lower-limb amputations who use prosthetic limbs. Materials and Methods: This non-experimental descriptive study assessed 58 persons with amputation who used prosthetic limbs. The mechanical axis angle (MAA) of the lower limb during the heel contact, midstance, and toe-off phases and the angle between the tube and floor during the midstance phase were measured using coronal plane gait images. We also investigated whether the MAA and tube angle during the midstance phase have a multimodal distribution. In case of multimodal distributions, we tested for significant between-group differences in patient characteristics. Results: The MAA and tube angle in the coronal plane during the midstance phase had a bimodal distribution (mean 0°). There was a significant difference in the duration of prosthetic limb use between the MAA Conclusions: In this study, prosthetic alignment during the midstance phase had a bimodal distribution. In both groups, deviations in the lower limb MAA were aligned to be approximately 3°.

Management of Anterior Tibialis Tendon Ruptures.

Anterior tibialis tendon ruptures, while relatively infrequent ruptures, are commonly identified in delayed fashion, which can lead to significant impairments in patient gait and function. Surgical treatment is typically required to restore ankle dorsiflexion function and proper gait. Depending on various patient-specific factors, tendon quality and excursion, and chronicity, a range of treatment options are available to manage these patients, from nonsurgical care to surgical treatment. Surgical options include direct repair, local tendon transfer, autograft tendon reconstruction, and allograft tendon reconstruction. Additional procedures may need to be considered. Despite the variety of described surgical procedures, limited evidence-based guidelines are available to direct surgeons in the most optimal treatment for their patients. In addition to the relevant anatomy, biomechanics, and pathoanatomy, the reconstructive armamentarium is detailed and reviewed here, along with outcomes and potential complications, to guide surgeons in the most appropriate treatment for their patients.

A Pioneering Physiotherapeutic Approach to the Treatment of a COVID Affected Patient – A Case Report

The Coronavirus (SARS-CoV-19) originated from Wuhan, China and has affected globe since 2019. The virus spread to India in January, 2020. It is highly contagious. In its severe form, the victims of the virus suffered from symptoms such as breathing difficulties, fever, weakness, loss of taste and smell, amongst others. Patients were put on mechanical ventilators and as a result, the requirement of the role of cardio-respiratory physiotherapists became more and more essential. Even after weaning off from the ventilators and being detected Coronavirus negative, the role of the physiotherapists still remained indispensable in order to bring about improvement in the muscle strength of the affected weakened muscles, especially for patients whose movements were limited for extended periods of times due to being dependent on the ventilators; physiotherapists were also required to bring about improvement in the lung capacities and function, to teach gait training for those who find it difficult to walk, to improve in-coordination and overall allow the patients to come back to normal and carry out all the ADL independently and well. He had a severe difficulty in breathing and coughing, along with a generalized weakness all over his body, specifically both his lower limbs. He was admitted in the ICU immediately and was on mechanical ventilator. After all the physiotherapy sessions, the patient was able to carry out his activities of daily living independently; he regained his muscle strength and improved his coordination and ability to walk with a proper gait. He was also able to breathe independently with adequate respiratory muscle strength; and did not feel fatigued as quickly as he used to when he was detected positive for Coronavirus.

Andreas Mayer, The Science of Walking: Investigations into Locomotion in the Long Nineteenth Century

In recent years, historians of science have made fruitful use of an analytical perspective focused on the definition of natural phenomena as ‘objects’ of science. Andreas Mayer turns this process of objectification on its head. Tracing the history of ‘a science that does not exist’, he examines ‘theories and practices of observation and experimentation’ (p.2) in anthropology, physiology, orthopedic surgery, neurology and psychiatry, as well as literature and the fine arts. Mayer concludes that the science of walking did not lead to a clearly defined research field with special laboratories, instruments and methods nor to the identification of a normalised form of walking. Rather, he points to ‘multifarious varieties of knowledge about the human gait’ that should be understood ‘as an interplay between mechanical, semiotic, and poetic registers’ (p.142). Mayer begins his account in the later eighteenth century, exploring Rousseau’s ‘anthropology of the solitary walker’ (p.10) and a travel literature that extolled the benefits of walking while decrying mechanical modes of transport and contraptions aimed at forming a proper gait. In this context, he takes up what becomes a persistent theme of the study, the role of the military in initiating ‘experiment-based investigation into the mechanics of gait’ (p.31). He also explores the beginnings of a ‘semiotic’ approach to styles of walking as seen in the physiognomy of J. C. Lavater and the work on mimicry of J. J. Engel, who asserted that ‘each specific mindset … has its own distinctive gait’ (p.36).

Symmetry and range of pelvic movement in gait among young male football players and their non-playing peers

Introduction: Gait is the basic physical activity of a person to which s/he is much better adapted than to running or maintaining a sitting position. Proper gait pattern has beneficial effects on locomotor and respiratory system efficacy, while asymmetrical, abnormal gait may cause musculoskeletal dysfunctions. The purpose of the study is to compare symmetry and range of pelvic movement in gait among boys playing football and their non-playing peers. Material and methods: A total of 121 boys were surveyed, including 33 soccer players aged 10-14 and 82 of their non-playing peers, using a portable system for functional movement analysis - BTS G-WALK. Symmetry and range of motion of the right and left pelvis in the sagittal, frontal and horizontal planes were analysed while walking on a straight section. Basic anthropometric measurements were also taken. Results: There was no difference in the value of the pelvic symmetry coefficient between the football- and non-players. However, a significantly greater range of pelvic rotation and movement in the sagittal plane was found among the boys playing football. Attention is drawn to the greater range of pelvic movement towards posterior tilt among young soccer players. Conclusions: Further studies should be conducted to assess whether excessive movement of the pelvis in gait towards posterior tilt among young soccer players is related to body posture quality, and whether the excessive range of pelvic movement in the sagittal and horizontal planes correlates with the incidence of locomotor injuries or lower spine pain.

Pedobarographic, Clinic, and Radiologic Evaluation after Surgically Treated Lisfranc Injury

Introduction Lisfranc injuries are rare, often missed, and may cause permanent structural deterioration of tarsometatarsal joint, despite optimal management. Consequently, a Lisfranc injury may lead to disruption of the biomechanics of the normal foot during walking and may alter the plantar pressure distribution, which is essential for proper gait mechanics. Therefore, the main purpose of the study was to specify the dynamic plantar pressure, radiographic and clinical features, after surgically managed Lisfranc injuries. Methods This study was carried out over a period of 10 years and included 62 patients who were surgically treated for Lisfranc injury, with mean 57-month follow-up. Radiological (intermetatarsal, Kite’s, first metatarsophalangeal, Meary’s, Hibbs’ and calcaneal pitch angles, and medial cuneiform–fifth metatarsal distance), pedobarographical, and clinical results with the American Orthopaedic Foot and Ankle Society (AOFAS) midfoot score assessments for both feet were assessed. Results In the radiological assessment, the mean first intermetatarsal (p = 0.006) and Meary’s angle (p = 0.000) were decreased on the injured feet compared to the uninjured feet on the anteroposterior and lateral radiographs. In the pedobarographic assessment, the injured feet midfoot contact time increased (p = 0.03), and maximum force (p = 0.001), total peak pressure (p = 0.008), and contact area (p = 0.017) decreased, compared to the uninjured feet. The mean AOFAS score was 75/100 at the final follow-up visit. There was seen to be reduced both contact surface area and time of the midfoot. Conclusion Despite surgical management of Lisfranc injuries, the injured foot does not regain functional, radiological, or pedobarographical levels as compared to the uninjured foot for ≥ 57 months.

Robotic hip-disarticulation prosthesis: evaluation of prosthetic gaits in a non-amputee individual

This is the first report regarding the application of a robotic prosthesis for hip-disarticulation amputees. Hip-disarticulation prostheses (HDPs) are used by amputees with either an amputation or anomaly at or close to the hip joint. Robotics can be beneficial in such cases, as these individuals have reduced levels of motor function compared to other amputees and have limited mobility. We developed a robotic HDP, and analysed the gaits of a non-amputee using the robotic and non-powered dummy HDPs. The results imply that a robotic HDP may induce proper gait and maintain the burdens of the prosthetic and non-prosthetic limbs symmetrically.

Design, development, and evaluation of a local sensor-based gait phase recognition system using a logistic model decision tree for orthosis-control

BackgroundFunctionality and versatility of microprocessor-controlled stance-control knee-ankle-foot orthoses (M-SCKAFO) are dictated by their embedded control systems. Proper gait phase recognition (GPR) is required to enable these devices to provide sufficient knee-control at the appropriate time, thereby reducing the incidence of knee-collapse and fall events. Ideally, the M-SCKAFO sensor system would be local to the thigh and knee, to facilitate innovative orthosis designs that allow more flexibility for ankle joint selection and other orthosis components. We hypothesized that machine learning with local sensor signals from the thigh and knee could effectively distinguish gait phases across different walking conditions (i.e., surface levels, walking speeds) and that performance would improve with gait phase transition criteria (i.e., current states depend on previous states).MethodsA logistic model decision tree (LMT) classifier was trained and tested (five-fold cross-validation) on gait data that included knee flexion angle, thigh-segment angular velocity, and thigh-segment acceleration. Twenty features were extracted from 0.1 s sliding windows for 30 able-bodied participants that walked on different surfaces (level, down-slope, up-slope, right cross-slope, left cross-slope) at a various walking speeds (self-paced (1.33 m/s, SD = 0.04 m/s), 0.8, 0.6, 0.4 m/s). The LMT-based GPR model was also tested with another validation set containing similar features and surfaces from 12 able-bodied volunteers at self-paced walking speeds (1.41 m/s, SD = 0.34 m/s). A “Transition Sequence Verification and Correction” (TSVC) algorithm was applied to correct for continuous class prediction and to improve GPR performance.ResultsThe LMT had a tree size of 1643 with 822 leaf nodes, with a logistic regression model at each leaf node. The local sensor LMT-based GPR model identified loading response, push-off, swing, and terminal swing phases with overall classification accuracy of 98.38 for the initial training set (five-fold cross-validation) and 90.60% for the validation set. Applying TSVC increased classification accuracy to 98.72% for the initial training set and 98.61% for the validation set. Sensitivity, specificity, precision, F-score, and Matthew’s correlation coefficient results suggest strong evidence for the feasibility of an LMT-based GPR system for real-time orthosis control.ConclusionsThe novel machine learning GPR model that used sensor features local to the thigh and knee was viable for dynamic knee-ankle-foot orthosis-control. This highly accurate GPR model was generalizable when combined with TSVC. Our approach could reduce sensor system complexity as compared with other M-SCKAFO approaches, thereby enabling customizable advantages for end-users through modular unit orthosis designs.

Differential associations between dual-task walking abilities and usual gait patterns in healthy older adults—Results from the Baltimore Longitudinal Study of Aging

BackgroundIt is well established that facing a cognitive challenge while carrying out a motor task interferes with the motor task performance, and in general the ability of handling a dual-task declines progressively with aging. However, the reasons for this decline have not been fully elucidated. Understanding the association between usual-walking gait patterns and dual-task walking performance may provide new insights into the mechanisms that lead to gait deterioration in normal aging and its link to motor and cognitive function. Research questionOur aim was to assess usual gait parameters in kinematics and kinetics to understand how these parameters are related with a specific task in dual-task walking. MethodsWe hypothesized that difficulty in dual-task walking would be associated with gait deteriorations as reflected in range of motion and mechanical work expenditure. We tested this hypothesis by quantifying the gait of 383 participants in the Baltimore Longitudinal Study of Aging (68% of whom successfully completed the dual-task walk, 21% failed the motor task, and 11% failed the cognitive task). ResultsCompared to successful performers, participants who failed the single motor task had slower gait speed, shorter stride length, higher cadence, and lower range of motion in the knee and ankle joints (p < 0.05, for all), while the participants who failed the cognitive task while walking had longer double support time (p = 0.003), and greater knee absorptive mechanical work (p = 0. 001) and lower ankle generative mechanical work (p < 0. 001). SignificanceThese results suggest that dual-task walking may be useful for monitoring subtle and diverse gait deteriorations in aging and possibly for designing interventions for maintaining and regaining proper gait patterns in older adults.