Search

Gait cycle variability during steady walking, described by the stride interval time series, has been used as a gait-stability-related measure. In particular, the positive persistency in the stride intervals with 1/f-like fluctuation and reduction of the persistency are the well-documented metrics that can characterize gait patterns of healthy young adults and elderly including patients with neurological diseases, respectively. Here, we examined effects of a dual task on gait cycle variability in healthy young adults, based on the mean and standard deviation statistics as well as the positive persistency of the stride intervals during steady walking on a treadmill. Specifically, three gait conditions were examined: control condition, non-cognitive task with holding a smartphone in front of the chest using their dominant hand and looking fixedly at a blank screen of the smartphone, and cognitive motor task with holding a smartphone as in the non-cognitive task and playing a puzzle game displayed on the smartphone by one-thumb operation. We showed that only the positive persistency, not the mean and standard deviation statistics, was affected by the cognitive and motor load of smartphone usage in the cognitive condition. More specifically, the positive persistency exhibited in the control and the non-cognitive conditions was significantly reduced in the cognitive condition. Our results suggest that the decrease in the positive persistency during the cognitive task, which might represent the deterioration of healthy gait pattern, is caused endogenously by the cognitive and motor load, not necessarily by the reduction of visual field as often hypothesized. [ABSTRACT FROM AUTHOR]

Hip joint prostheses are used to replace hip joint function in the human body. The latest dual-mobility hip joint prosthesis has an additional component of an outer liner that acts as a cover for the liner component. Research on the contact pressure generated on the latest model of a dual-mobility hip joint prosthesis under a gait cycle has never been done before. The model is made of ultrahigh molecular weight polyethylene (UHMWPE) on the inner liner and 316L stainless steel (SS 316L) on the outer liner and acetabular cup. Simulation modeling using the finite element method is considered static loading with an implicit solver for studying the geometric parameter design of dual-mobility hip joint prostheses. In this study, simulation modeling was carried out by applying varying inclination angles of 30°, 40°, 45°, 50°, 60°, and 70° to the acetabular cup component. Three-dimensional loads were placed on femoral head reference points with variations of femoral head diameter used at 22 mm, 28 mm, and 32 mm. The results in the inner surface of the inner liner, the outer surface of the outer liner, and the inner surface of the acetabular cup showed that the variations in inclination angle do not have a major effect on the maximum contact pressure value on the liner component, where the acetabular cup with an inclination angle of 45° can reduce contact pressure more than the other studied inclination angle variations. In addition, it was found that the 22 mm diameter of the femoral head increases the contact pressure. The use of a larger diameter femoral head with an acetabular cup configuration at a 45° inclination can minimize the risk of implant failure due to wear. [ABSTRACT FROM AUTHOR]

Pelvic bone is a complex and robust load-bearing skeletal structure in the human body, the evolution of which might have been influenced by mechanical loads of daily activities like walking, upright standing, and running. Since the main function of skeletal bones is to provide rigidity to the body and provide hard surfaces for muscle attachment as well, in this work we propose a compliance minimization problem to determine whether material distribution guided by topology optimization yields a skeletal structure similar to the pelvic bone under same boundary conditions and volumetric constraints. As bone growth occurs in response to the mechanical loads acting on it, we consider the maximal loads that the pelvic bone may experience for a continued period of time, namely during running. The running gait cycle is divided into seven phases, and the objective function is a weighted combination of these seven phases. The optimal geometries are compared with the natural hemi-pelvis by measuring shape similarity using Procrustes analysis. Results show that the optimal geometries have good shape similarity in stance phases. We also explore the design space by considering a combination of sequence of phases which is an alternative to the weighted multiple load-case objective function. In all cases, the optimal geometries are stiffer than the hip bone. To validate this result, we conducted compression test experiments on selected optimal geometries and natural hemi-pelvis model of same material and found that the experimental results prove that topology optimization based optimal geometries are indeed more stiff than the natural hemi-pelvis geometry. [ABSTRACT FROM AUTHOR]

Most conventional biped robots process leg movements and information from each sensor by numerical calculation using a CPU. However, to cope with diverse environments, the numerical calculations are enormous, so they must be processed at high speed using a high-performance CPU and high power consumption. On the other hand, focusing on human motor control, it is believed that basic motor patterns such as walking and running are generated by a neural network called the central pattern generator (CPG), which is localized in the spinal cord and is independent of calculation. We previously focused on pulse-type hardware neural networks (P-HNNs), in which the neural network was composed of analog electronic circuits, and developed a hardware CPG model for controlling a single leg of a musculoskeletal humanoid robot. However, to actually move a biped robot, a CPG model that takes into account both legs and sensory information is required. Therefore, this study aims to develop a hardware CPG model for controlling both legs of a musculoskeletal humanoid robot whose gait changes according to the higher center and sensory information. We report on a hardware CPG model configured by circuit simulation confirmed the generation of walking and running patterns. [ABSTRACT FROM AUTHOR]

A Finite Element Analysis (FEA) was performed to evaluate the interaction between residual limb and socket when considering the dynamic loads of the gait cycle. Fourteen transfemoral amputees participated in this study, where their residual limbs (i.e., soft tissues and bone), and their sockets were reconstructed. The socket and the femur were defined as elastic materials, while the bulk soft tissues were defined as a hyperelastic material. Each model included the donning, standing, and gait cycle phase, with load and boundary conditions applied accordingly. The influence of adding the dynamic loads related to the gait cycle were compared against the modelling of the static load equivalent to the standing position resulting in changes of 23% ± 19% in the maximum values and in an increase in the size of the regions where they were located. Additionally, the possible correspondence between comfort and the location of peak loadbearing at the residual-limb/socket interface was explored. Consequently, the comfort perceived by the patient could be estimated based on the locations of the maximum stresses (i.e., if they coincide with the pressure tolerant or sensitive regions of the residual limb). [ABSTRACT FROM AUTHOR]

To reduce breast motion with a bra, we need to understand what drives the motion of the breasts, and what variables change as support increases. Quantifying breast-torso coordination and movement complexity across the gait cycle may offer deeper insights than previously reported discrete time lag. We aimed to compare breast-torso coordination and mutual influence across breast support conditions during running. Twelve female participants ran on a treadmill at 10 km h−1 with an encapsulation and compression sports bra, and in no bra. Nipple and torso position was recorded. Vector coding, granger causality and transfer entropy were calculated within gait cycles. In both bra conditions, a greater percentage of gait cycles was spent with the breast and torso in-phase (> 90%) compared to no bra running (~ 66%, p < 0.001), with most time spent in-phase in the encapsulation versus compression bra (p = 0.006). There was a main effect of breast support condition on Granger causality (p < 0.001), both from breast to torso and torso to breast. Transfer of information was highest from torso to breast, compared to breast to torso in all conditions. Overall, these results provide novel insight into the mutual and complex interaction between the breast and the torso while running in different bra conditions. The approaches presented allow for a greater understanding of bra support conditions than existing discrete measures, which may relate to comfort and performance. Therefore, measures of coupling, predictability and transfer of complexity should be employed in future work examining these features. [ABSTRACT FROM AUTHOR]

This study evaluated gait variability in patients before and after total knee arthroplasty (TKA) using the equality of variance method to determine where variability differences occur in the movement cycle. Twenty-eight patients underwent TKA with cruciate-sacrificed implants. Patients underwent motion analysis which measured knee biomechanics as they walked overground at their preferred pace before and 12 months after TKA. Equality of variance results were compared with 14 healthy controls of similar age. Before surgery, patients had reduced knee extension moment variability throughout the early stance phase (4–21% gait cycle, p < 0.05) compared to controls. Knee power variability was lower preoperatively compared to controls for most of the stance phase (0–13% and 17–60% gait cycle, p < 0.05). Sagittal knee moment and power variability further decreased following TKA. Knee extension moment variability was lower postoperatively throughout stance phase compared to preoperatively (4–22% and 36–60% gait cycle, p < 0.05) and compared to controls (4–30% and 45–60% gait cycle, p < 0.05). Knee power variability remained lower following TKA throughout stance phase compared to preoperatively (10–24% and 36–58% gait cycle, p < 0.05) and controls (3–60% gait cycle, p < 0.05). TKA patients may be less stable, and this may be in part due to an unresolved adaptation developed while awaiting TKA surgery and the cruciate sacrificing design of the implants utilized in this study. [ABSTRACT FROM AUTHOR]

Electromyography (EMG) is considered a potential predictive tool for the severity of knee osteoarthritis (OA) symptoms and functional outcomes. Patient-reported outcome measures (PROMs), such as the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and visual analog scale (VAS), are used to determine the severity of knee OA. We aim to investigate muscle activation and co-contraction patterns through EMG from the lower extremity muscles of patients with advanced knee OA patients and evaluate the effectiveness of an interpretable machine-learning model to estimate the severity of knee OA according to the WOMAC (pain, stiffness, and physical function) and VAS using EMG gait features. To explore neuromuscular gait patterns with knee OA severity, EMG from rectus femoris, medial hamstring, tibialis anterior, and gastrocnemius muscles were recorded from 84 patients diagnosed with advanced knee OA during ground walking. Muscle activation patterns and co-activation indices were calculated over the gait cycle for pairs of medial and lateral muscles. We utilized machine-learning regression models to estimate the severity of knee OA symptoms according to the PROMs using muscle activity and co-contraction features. Additionally, we utilized the Shapley Additive Explanations (SHAP) to interpret the contribution of the EMG features to the regression model for estimation of knee OA severity according to WOMAC and VAS. Muscle activity and co-contraction patterns varied according to the functional limitations associated with knee OA severity according to VAS and WOMAC. The coefficient of determination of the cross-validated regression model is 0.85 for estimating WOMAC, 0.82 for pain, 0.85 for stiffness, and 0.85 for physical function, as well as VAS scores, utilizing the gait features. SHAP explanation revealed that greater co-contraction of lower extremity muscles during the weight acceptance and swing phases indicated more severe knee OA. The identified muscle co-activation patterns may be utilized as objective candidate outcomes to better understand the severity of knee OA. [ABSTRACT FROM AUTHOR]

Lower limb amputees face significant challenges in day-to-day life for locomotion. A prosthetic foot is used to restore mobility. The push-off effect from the prosthetic foot plays a vital role as it reduces gait asymmetry and sound limb loading. In the present work, four prosthetic foot, two conventional prosthetic foot (Jaipur foot and Ranger foot) and two energy-storing and return prosthetic foot (Nylon foot and Carbon fibre foot) are investigated for gait parameters of the unilateral transtibial amputee. Also, a novel wearable setup is used to acquire the amputee data. It comprises of two rotary potentiometers at the knee joint and two wearable slippers with four force sensitive resistor sensors on each. The gait parameters investigated in this study are the range of motion (ROM) of the knee joint, loading force on limbs, spatial and temporal gait parameters. Stability analysis between amputated and sound limbs is analyzed by measuring gait symmetry for gait parameters. Also, the coefficient of variation is calculated to see how gait parameters vary with the change of prosthetic foot. ROM symmetry and loading force symmetry is obtained maximum for Carbon fibre foot whereas minimum ROM symmetry is observed for Ranger foot. Furthermore, Carbon fibre foot attains maximum loading force symmetry of 0.98 indicating maximum push-off among all the four selected foot. [ABSTRACT FROM AUTHOR]

Joint kinematics are an important and widely utilized metric in quantitative human movement analysis. Typically, trajectory data for skin-mounted markers are collected using stereophotogrammetry, sometimes referred to as optical motion capture, and processed using various mathematical models to estimate joint kinematics (e.g., angles). Among the various sources of noise in optical motion capture data, soft tissue artifacts (STAs) remain a critical source of error. This study investigates the performance of the point cluster technique (PCT), an extension of the PCT using perturbation theory (PCT-PT), and singular value decomposition least squares (SVD-LS) method (as a reference) for 100 different marker configurations on the thigh and shank during treadmill walking. This study provides additional evidence that the PCT method is significantly limited by the underlying mathematical constraints governing its optimization process. Furthermore, the results suggest the PCT-PT method outperforms the PCT method across all performance metrics for both body segments during the entire gait cycle. For position-based metrics, the PCT-PT method provides better estimates than the SVD-LS method for the thigh during majority of the stance phase and provides comparable estimates for the shank during the entire gait cycle. For knee angle estimates, the PCT-PT method provides equivalent results as the SVD-LS method. [ABSTRACT FROM AUTHOR]

Medial meniscus extrusion (MME) is exacerbated by repeated mechanical stress. Various factors would affect MME; however, there is limited information about the behaviour of the medial meniscus during walking in patients with knee osteoarthritis (KOA). This study aimed to investigate the pattern of MME during walking and its association with limb biomechanics in patients with KOA. Fifty-five patients with KOA and ten older adult volunteers as a control group were involved in this study. The MME and limb biomechanics during walking were evaluated simultaneously by ultrasound and a motion analysis system, respectively. The waveform was constructed from the values of MME, and the point showing the highest value of MME was identified during the gait cycle. According to the peak timing of MME in the waveform, the pattern of the waveform was evaluated and compared to the control group. Lateral thrust, knee adduction moment (KAM), and flexion moment were obtained from motion analysis, and their association with the MME was evaluated. The patients with KOA demonstrated unique peak timing during walking. Compared to the control group, there were three groups of MME waveforms, early (< 59%), normal (60–83%), and late (> 84%) from the peak timing in the gait cycle. The pattern of MME waveform in early, normal, and late groups was correlated with the first KAM and lateral thrust, second KAM, and knee flexion moment, respectively. A unique MME pattern during walking was demonstrated, and these patterns were associated with limb biomechanics in patients with KOA. [ABSTRACT FROM AUTHOR]

Walking and typing on a smartphone is an extremely common interaction. Previous research has shown that error rates are higher when walking than when stationary. In this paper we analyse the acceleration data logged in an experiment in which users typed whilst walking, and extract the gait phase angle. We find statistically significant relationships between tapping time, error rate and gait phase angle. We then use the gait phase as an additional input to an offset model, and show that this allows more accurate touch interaction for walking users than a model which considers only the recorded tap position. [ABSTRACT FROM AUTHOR]

The current primary treatment approach for malignant pelvic tumors involves hemipelvic prosthesis reconstruction following tumor resection. In cases of Enneking type II + III pelvic tumors, the prosthesis necessitates fixation to the remaining iliac bone. Prevailing methods for prosthesis fixation include the saddle prosthesis, ice cream prosthesis, modular hemipelvic prosthesis, and personalized prosthetics using three-dimensional printing. To prevent failure of hemipelvic arthroplasty protheses, a novel fixation method was designed and finite element analysis was conducted. In clinical cases, the third and fourth sacral screws broke, a phenomenon also observed in the results of finite element analysis. Based on the original surgical model, designs were created for auxiliary dorsal iliac, auxiliary iliac bottom, auxiliary sacral screw, and auxiliary pubic ramus fixation. A nonlinear quasi-static finite element analysis was then performed under the maximum load of the gait cycle, and the results indicated that assisted sacral dorsal fixation significantly reduces stress on the sacral screws and relative micromotion exceeding 28 μm. The fixation of the pubic ramus further increased the initial stability of the prosthesis and its interface osseointegration ability. Therefore, for hemipelvic prostheses, incorporating pubic ramus support and iliac back fixation is advisable, as it provides new options for the application of hemipelvic tumor prostheses. [ABSTRACT FROM AUTHOR]

Gait analysis is often used to study locomotor alterations in people with multiple sclerosis (PwMS), but the large number of extracted variables challenges the interpretability. In this paper, we analysed gait alterations by combining the Gait Profile Score (GPS), which summarizes kinematic locomotor deviations, and Statistical Parametric Mapping (SPM), which compares kinematics and kinetics over the whole gait cycle. Eleven PwMS and 11 speed-matched Healthy Controls (HC) underwent overground gait analysis. GPS were compared through independent-samples t-tests; sagittal-plane kinematics and power at hip, knee, and ankle were compared through SPM Hotelling's-T2 and SPM t-tests. Spearman's correlation coefficients (r) between GPS and clinical outcomes were also calculated. PwMS had higher GPS than HC (PwMS = 8.74 ± 2.13°; HC = 5.01 ± 1.41°;p < 0.001). Multivariate SPM found statistically significant differences at 0–49%, 70–80%, and 93–99% of stride (p < 0.05) and univariate analysis showed reduced ankle dorsiflexion, and lower knee flexion during pre-swing and swing. GPS correlated with Expanded Disability Status Scale (r = 0.65; 95%C.I.[0.04,0.91]; p = 0.04) and 2-Minute Walking Test (r = -0.65; 95%C.I.[-0.91,-0.04]; p = 0.04). GPS in conjunction with SPM revealed multi-joint kinematic alterations on sagittal plane involving distal joint angles, ankle and knee, during the stance phase with no changes at the proximal level. Gait deviations were more pronounced in PwMS with higher disability and walking limitations. [ABSTRACT FROM AUTHOR]

In human walking, whole-body angular momentum (WBAM) about the body centre-of-mass is reportedly maintained in a small range throughout a gait cycle by the intersegmental cancellation of angular momentum. However, the WBAM is certainly not zero, which indicates that external moments applied from the ground due to ground reaction forces (GRFs) and vertical free moments (VFMs) counteract the WBAM. This study provides a complete dataset of the WBAM, each segmental angular momentum, and the external moments due to GRFs and VFMs during human walking. This is done to test whether (1) the three components of the WBAM are cancelled by coordinated intersegmental movements, and whether (2) the external moments due to GRFs and VFMs contribute only minimally to WBAM regulation throughout a gait cycle. This study demonstrates that WBAM is regulated in a small range not only by the segment-to-segment cancellation, but also largely through contributions by the GRFs. The magnitude of VFM is significantly smaller than the peak vertical moment generated by the GRFs; however, in the single-support phase during walking, the VFM is possibly critical for coping with the change in the vertical WBAM due to force perturbations and arm or trunk movements. [ABSTRACT FROM AUTHOR]

Quantitative Gait Analysis (QGA) is considered as an objective measure of gait performance. In this study, we aim at designing an artificial intelligence that can efficiently predict the progression of gait quality using kinematic data obtained from QGA. For this purpose, a gait database collected from 734 patients with gait disorders is used. As the patient walks, kinematic data is collected during the gait session. This data is processed to generate the Gait Profile Score (GPS) for each gait cycle. Tracking potential GPS variations enables detecting changes in gait quality. In this regard, our work is driven by predicting such future variations. Two approaches were considered: signal-based and image-based. The signal-based one uses raw gait cycles, while the image-based one employs a two-dimensional Fast Fourier Transform (2D FFT) representation of gait cycles. Several architectures were developed, and the obtained Area Under the Curve (AUC) was above 0.72 for both approaches. To the best of our knowledge, our study is the first to apply neural networks for gait prediction tasks. [ABSTRACT FROM AUTHOR]

Human identification using unobtrusive visual features is a daunting task in smart environments. Gait is among adequate biometric features when the camera cannot correctly capture the human face due to environmental factors. In recent years, gait-based human identification using skeleton data has been intensively studied using a variety of feature extractors and more sophisticated deep learning models. Although skeleton data is susceptible to changes in covariate variables, resulting in noisy data, most existing algorithms employ a single feature extraction technique for all frames to generate frame-level feature maps. This results in degraded performance and additional features, necessitating increased computing power. This paper proposes a robust feature extractor that extracts a quantitative summary of gait event-specific information, thereby reducing the total number of features throughout the gait cycle. In addition, a novel Attention-guided LSTM-based deep learning model with residual connections is proposed to learn the extracted features for gait recognition. The proposed approach outperforms the state-of-the-art works on five publicly available datasets on various benchmark evaluation protocols and metrics. Further, the CMC test revealed that the proposed model obtained higher than 97% Accuracy in lower-level ranks on these datasets. [ABSTRACT FROM AUTHOR]

Gait is biometric that refers to the walking style of an individual, and medical studies claim that every individual has a distinctive gait pattern. In this work, we consider a gait recognition scenario in which at least a half cycle of gait is captured from the fronto-parallel view for each person and propose a new feature termed the Pose-based Boundary Energy Image that captures the dynamics of gait by extracting features from the silhouette contour information corresponding to several fractional parts of a gait cycle. Traditionally, binary silhouettes extracted from the input RGB frames of a gait sequence were used to construct the gait features for recognition. In contrast, silhouette contour information for feature extraction, as used in this work, is advantageous in the sense that it makes the extracted gait features more discriminative due to eliminating the redundant pixel-level structural information. We next reduce the dimension of the extracted Pose-based BEI feature using PCA, and finally carry out LDA-based classification using the reduced feature set. Evaluation of the proposed approach has been done using the popular CASIA-B and TUMGAID data sets and quite satisfactory results are obtained. A comparative study with existing techniques also shows that our approach outperforms the existing methods both in terms of average accuracy and classification time. [ABSTRACT FROM AUTHOR]

Digital quantification of gait can be used to measure aging- and disease-related decline in mobility. Gait performance also predicts prognosis, disease progression, and response to therapies. Most gait analysis systems require large amounts of space, resources, and expertise to implement and are not widely accessible. Thus, there is a need for a portable system that accurately characterizes gait. Here, depth video from two portable cameras accurately reconstructed gait metrics comparable to those reported by a pressure-sensitive walkway. 392 research participants walked across a four-meter pressure-sensitive walkway while depth video was recorded. Gait speed, cadence, and step and stride durations and lengths strongly correlated (r > 0.9) between modalities, with root-mean-squared-errors (RMSE) of 0.04 m/s, 2.3 steps/min, 0.03 s, and 0.05–0.08 m for speed, cadence, step/stride duration, and step/stride length, respectively. Step, stance, and double support durations (gait cycle percentage) significantly correlated (r > 0.6) between modalities, with 5% RMSE for step and stance and 10% RMSE for double support. In an exploratory analysis, gait speed from both modalities significantly related to healthy, mild, moderate, or severe categorizations of Charleson Comorbidity Indices (ANOVA, Tukey's HSD, p < 0.0125). These findings demonstrate the viability of using depth video to expand access to quantitative gait assessments. [ABSTRACT FROM AUTHOR]

Knee osteoarthritis (OA) and obesity are major public health concerns that are closely intertwined. This intimate relationship was documented by considering obesity as the most significant preventable risk factor associated with knee OA. To date, however, the effects of obesity on the knee joint's passive-active structure and cartilage loading have been inconclusive. Hence, this study investigates the intricate relationship between obesity and knee OA, centering on the biomechanical changes in knee joint active and passive reactions during the stance phase of gait. Using a subject-specific musculoskeletal and finite element approach, muscle forces, ligament stresses, and articular cartilage contact stresses were analyzed among 60 individuals with different body mass indices (BMI) classified under healthy weight, overweight, and obese categories. Our predicted results showed that obesity significantly influenced knee joint mechanical reaction, increasing muscle activations, ligament loading, and articular cartilage contact stresses, particularly during key instances of the gait cycle—first and second peak loading instances. The study underscores the critical role of excessive body weight in exacerbating knee joint stress distribution and cartilage damage. Hence, the insights gained provide a valuable biomechanical perspective on the interaction between body weight and knee joint health, offering a clinical utility in assessing the risks associated with obesity and knee OA. [ABSTRACT FROM AUTHOR]

Background: Pes planovalgus is one of the most common pediatric skeletal deformities. There have been no studies to analyze in detail the spatiotemporal variables of gait following arthroereisis. Purpose of the study: The purpose of our study was to assess gait parameters in patients with symptomatic flexible flatfoot following treatment with the talus screw. Methods: This was a prospective study assessing the 22 patients treated surgically due to symptomatic flexible flatfoot with the talus screw. Patients underwent gait assessment with a G-Sensor. We analyzed the following gait parameters: gait cycle duration, step length, support phase duration, swing phase duration, double support duration, single support duration, cadence, velocity, step length. Results: The post-operative gait parameter assessment for the operated and non-operated foot showed a significant difference only in terms of step length. Cadence increased from the pre-operative mean of 82.29 steps/min to a post-operative mean of 102.94 steps/min. Gait velocity increased significantly from 0.81 m/s before to 0.96 m/s after surgery. Discussion: Arthroereisis with the talus screw helps improve gait parameters following surgery. Post-operatively, we observed increased gait velocity and cadence and decreased gait cycle duration in the operated limb. Conclusion: Short-term biomechanical outcomes of pes planovalgus treatment with the talus screw are good. [ABSTRACT FROM AUTHOR]

The article focuses on enhancing lower limb exoskeletons for various terrains through a new approach using reinforcement learning. Topics discussed include the challenges of smooth transitions between different locomotion modes, the use of generative models like Variational Autoencoders (VAEs) for gait cycle representation, and the application of reinforcement learning to improve gait trajectory and event predictions.

Trajectory generation for biped robots is very complex due to the challenge posed by real-world uneven terrain. To address this complexity, this paper proposes a data-driven Gait model that can handle continuously changing conditions. Data-driven approaches are used to incorporate the joint relationships. Therefore, the deep learning methods are employed to develop seven different data-driven models, namely DNN, LSTM, GRU, BiLSTM, BiGRU, LSTM+GRU, and BiLSTM+BiGRU. The dataset used for training the Gait model consists of walking data from 10 able subjects on continuously changing inclines and speeds. The objective function incorporates the standard error from the inter-subject mean trajectory to guide the Gait model to not accurately follow the high variance points in the gait cycle, which helps in providing a smooth and continuous gait cycle. The results show that the proposed Gait models outperform the traditional finite state machine (FSM) and Basis models in terms of mean and maximum error summary statistics. In particular, the LSTM+GRU-based Gait model provides the best performance compared to other data-driven models. [ABSTRACT FROM AUTHOR]

Lower-limb assisted exoskeletons are widely researched for movement assistance or rehabilitation training. Due to advantages of compliance with human body and lightweight, some cable-driven prototypes have been developed, but most of these can assist only unidirectional movement. In this paper we present an untethered cable-driven ankle exoskeleton that can achieve plantarflexion-dorsiflexion bidirectional motion bilaterally using a pair of single motors. The main weights of the exoskeleton, i.e., the motors, power supplement units, and control units, were placed close to the proximity of the human body, i.e., the waist, to reduce the redundant rotation inertia which would apply on the wearer's leg. A cable force transmission system based on gear-pulley assemblies was designed to transfer the power from the motor to the end-effector effectively. A cable self-tension device on the power output unit was designed to tension the cable during walking. The gait detection system based on a foot pressure sensor and an inertial measurement unit (IMU) could identify the gait cycle and gait states efficiently. To validate the power output performance of the exoskeleton, a torque tracking experiment was conducted. When the subject was wearing the exoskeleton with power on, the muscle activity of the soleus was reduced by 5.2% compared to the state without wearing the exoskeleton. This preliminarily verifies the positive assistance effect of our exoskeleton. The study in this paper demonstrates the promising application of a lightweight cable-driven exoskeleton on human motion augmentation or rehabilitation. [ABSTRACT FROM AUTHOR]

Despite the ability of 3D convolutional methods to extract spatio-temporal information simultaneously, they also increase parameter redundancy and computational and storage costs. Previous work that has utilized the 2D convolution method has approached the problem in one of two ways: either using the entire body sequence as input to extract global features or dividing the body sequence into several parts to extract local features. However, global information tends to overlook detailed information specific to each body part, while local information fails to capture relationships between local regions. Therefore, this study proposes a new framework for constructing spatio-temporal representations, which involves extracting and fusing features in a novel manner. To achieve this, we introduce the multi-feature extraction-fusion (MFEF) module, which includes two branches: each branch extracts global features or local features individually, after which they are fused using multiple strategies. Additionally, as gait is a periodic action and different body parts contribute unequally to recognition during each cycle, we propose the periodic temporal feature modeling (PTFM) module, which extracts temporal features from adjacent frame parts during the complete gait cycle, based on the fused features. Furthermore, to capture fine-grained information specific to each body part, our framework utilizes multiple parallel PTFMs to correspond with each body part. We conducted a comprehensive experimental study on the widely used public dataset CASIA-B. Results indicate that the proposed approach achieved an average rank-1 accuracy of 97.2% in normal walking conditions, 92.3% while carrying a bag during walking, and 80.5% while wearing a jacket during walking. [ABSTRACT FROM AUTHOR]

Particularities in the individuals' style of walking have been explored for at least three decades as a biometric trait, empowering the automatic gait recognition field. Whereas gait recognition works usually focus on improving end-to-end performance measures, this work aims at understanding which individuals' traces are more relevant to improve subjects' separability. For such, a manifold projection technique and a multi-sensor gait dataset were adopted to investigate the impact of each data source characteristics on this separability. Assessments have shown it is hard to distinguish individuals based only on their walking patterns in a subject-based identification scenario. In this setup, the subjects' separability is more related to their physical characteristics than their movements related to gait cycles and biomechanical events. However, this study's results also points to the feasibility of learning identity characteristics from individuals' walking patterns learned from similarities and differences between subjects in a verification setup. The explorations concluded that periodic components occurring in frequencies between 6 and 10 Hz are more significant for learning these patterns than events and other biomechanical movements related to the gait cycle, as usually explored in the literature. [ABSTRACT FROM AUTHOR]

Exoskeletons aim to provide required torques in a patient’s lower limbs generating anatomical patterns. In this sense, this work presents an AGoRA lower limb exoskeleton aimed at rehabilitating patients who suffer hemiparesis. Two control strategies for physical Human-Robot Interaction are tested in a pilot study with a healthy user. On the one hand, the assistance mode uses a gait phase detection to generate a desired gait cycle in the paretic limb. On the other hand, the transparency mode is presented to simulate a back-drivable device using a mass-damper system. As a result, in the assistance mode, torques in the range of (20 - 30 Nm) for the knee and hip joints were generated to complete the user’s gait cycle. Moreover, the user can execute unrestricted movements using the AGoRA exoskeleton in the transparency mode. [ABSTRACT FROM AUTHOR]

This paper presents an automatic technique to detect and track the multiple pedestrians for their identifications in a video sequence. Contrarily to the most existing approaches, the proposed technique does not require human silhouette segmentation from the video to build the gait representation. Additionally, it also does not need to estimate the gait cycle to compute the gait-related features. The proposed technique comprises on four steps. In the first step, the pedestrian information is detected and tracked in the temporal direction. Second, we computed spatiotemporal features in the localized/tracked area to encode their walking patterns using dense trajectories. In the third step, the local features of pedestrian's walk are transformed into its compact and high-level representation using Fisher vector encoding scheme. Fourth, these high-level representations are fed to simple linear support vector machine for the identification. Since there is no publicly available multi-subject gait dataset and the recording of a new dataset is an expensive process which also demands a long time, we generated an augmented gait dataset where multiple subjects are available in a video sequence to cope with this limitation. We employed the single-subject CASIA-B gait dataset to generate the augmented multi-subject gait video sequences. The identification of multiple pedestrians in the constructed augmented gait sequences is a challenging task as multiple subjects are walking beside and crossing each other, hence producing several types of occlusions. The proposed gait recognition algorithm achieved a recognition rate of 86.3% on multi-subject gait dataset and 98.6% on the single-subject gait dataset. [ABSTRACT FROM AUTHOR]

Walking is an everyday activity and contains variations from person to person, from one step to another step. The variation may occur due to the uniqueness of each gait cycle, personal parameters, such as age, walking speed, etc., and the existence of a gait abnormality. Understanding the normal variation depending on personal parameters helps medical experts to identify deviations from normal gait and engineers to design compatible orthotic and prosthetic products. In the present study, we aimed to obtain normal gait variations based on age, sex, height, weight, and walking speed. For this purpose, a large dataset of walking trials was used to model normal walking. An artificial neural network-based gait characterization model is proposed to show the relation between personal parameters and gait parameters. The neural network model simulates normal walking by considering the effect of personal parameters. The predicted behavior of gait parameters by artificial neural network model has a similarity with existing literature. The differences between experimental data and the neural network model were calculated. To determine how much deviation between predictions and experiments can be considered excessive, the distributions of differences for each gait parameter were obtained. The phases of walking in which excessive differences were intensified were determined. It was revealed that the artificial neural network-based gait characterization model exhibits the behavior of the normal gait parameters depending on the personal parameters. [ABSTRACT FROM AUTHOR]

This data descriptor introduces GaitRec-VR, a 3D gait analysis dataset consisting of 20 healthy participants (9 males, 11 females, age range 21–56) walking at self-selected speeds in a real-world laboratory and the virtual reality (VR) replicas of this laboratory. Utilizing a head-mounted display and a 12-camera motion capture system alongside a synchronized force plate, the dataset encapsulates real and virtual walking experiences. A direct kinematic model and an inverse dynamic approach were employed for kinematics and computation of joint moments respectively, with an average of 23 ± 6 steps for kinematics and five clean force plate strikes per participant for kinetic analysis. GaitRec-VR facilitates a deeper understanding of human movement in virtual environments, particularly focusing on dynamic balance during walking in healthy adults, crucial for effective VR applications in clinical settings. The dataset, available in both.c3d and.csv formats, allows further exploration into VR's impact on gait, bridging the gap between physical and virtual locomotion. [ABSTRACT FROM AUTHOR]

Gait-type recognition is important for robotic exoskeletons and walking-assistance devices to adjust their output according to the users' needs. However, the growing trend of using machine learning (ML) models is both labor- and data-intensive, which makes it practically less attractive for application in exoskeletons and wearable-assistive devices. This research aims to devise a fuzzy-based gait recognition algorithm that requires minimum training data (only 40 cycles for each of the 5 gait types) and adapts to new users without having the need of pre-training for each of them. The proposed algorithm uses the fuzzy logic system (FLS) and Welford's (variance computation) method to enhance the adaptability by adjusting the rules for gait-type recognition and fine-tuning them in real time for every new user without requiring a specific prior training. Simulation-based evaluation of the proposed algorithm shows a gait-type recognition accuracy of 63.0%, an improvement of 36.8% over the non-adaptive fuzzy-based recognition algorithm. Moreover, the results show that the proposed algorithm outperforms the popular ML methods (support vector machine, Naive Bayes classifier, and logistic regression) when subjected to limited gait-cycles data and no prior training is provided. [ABSTRACT FROM AUTHOR]

Late-life balance disorders remain a severe problem with fatal consequences. Perturbation-based balance training (PBT), a form of rehabilitation that intentionally introduces small, unpredictable disruptions to an individual's gait cycle, can improve balance. The Tethered Pelvic Assist Device (TPAD) is a cable-driven robotic trainer that applies perturbations to the user's pelvis during treadmill walking. Earlier work showcased improved gait stability and the first evidence of increased cognition acutely. The mobile Tethered Pelvic Assist Device (mTPAD), a portable version of the TPAD, applies perturbations to a pelvic belt via a posterior walker during overground gait, as opposed to treadmill walking. Forty healthy older adults were randomly assigned to a control group (CG, n = 20) without mTPAD PBT or an experimental group (EG, n = 20) with mTPAD PBT for a two-day study. Day 1 consisted of baseline anthropometrics, vitals, and functional and cognitive measurements. Day 2 consisted of training with the mTPAD and post-interventional cognitive and functional measurements. Results revealed that the EG significantly outperformed the CG in several cognitive (SDMT-C and TMT-B) and functional (BBS and 4-Stage Balance: one-foot stand) measurements while showcasing increased confidence in mobility based on FES-I. To our knowledge, our study is the first randomized, large group (n = 40) clinical study exploring new mobile perturbation-based robotic gait training technology. [ABSTRACT FROM AUTHOR]

Vestibular balance control is dynamically weighted during locomotion. This might result from a selective suppression of vestibular inputs in favor of a feed-forward balance regulation based on locomotor efference copies. The feasibility of such a feed-forward mechanism should however critically depend on the predictability of head movements (HMP) during locomotion. To test this, we studied in 10 healthy subjects the differential impact of a stochastic vestibular stimulation (SVS) on body sway (center-of-pressure, COP) during standing and walking at different speeds and compared it to activity-dependent changes in HMP. SVS-COP coupling was determined by correlation analysis in frequency and time domains. HMP was quantified as the proportion of head motion variance that can be explained by the average head trajectory across the locomotor cycle. SVS-COP coupling decreased from standing to walking and further dropped with faster locomotion. Correspondingly, HMP increased with faster locomotion. Furthermore, SVS-COP coupling depended on the gait-cycle-phase with peaks corresponding to periods of least HMP. These findings support the assumption that during stereotyped human self-motion, locomotor efference copies selectively replace vestibular cues, similar to what was previously observed in animal models. [ABSTRACT FROM AUTHOR]

Background: The aim of the present study was to identify gait parameters in elderly Japanese women. 30 elderly women (65-74.9 years old) and 19 very elderly women (≥75 years old) participated in this study. A 3-dimensional (3D) motion analysis system was used to collect kinematic data, and a total of 70 gait parameters were analysed. Gait parameters included basic gait parameters, gait cycle parameters, and joints angle parameters, as well as angular velocity parameters, such as peak velocity and timing at the hip, knee, and ankle joints. Results: Our results indicated that basic gait parameters, such as the gait cycle, peak joint angle timing, and angular velocity parameters, significantly differ between elderly and very elderly women. Delayed peak joint angle timing and angular velocity parameters occurred during critical phases throughout the gait cycle: pre-swing, initial swing, and terminal swing phases. Conclusions: Several gait parameters exhibited significant differences between elderly and very elderly women. The timing of the peak joint angle and angular velocity parameters are primary characteristics defining gait changes in the elderly. [ABSTRACT FROM AUTHOR]

In this paper, an interpretable classifier for Parkinson's Disease (PD) diagnosis based on analyzing the gait cycle is presented. The proposed method utilizes clinical features extracted from the vertical Ground Reaction Force (vGRF) measured by wearable sensors. Therefore, experts can verify the decision made by the proposed method. Type-2 fuzzy logic is applied to increase the robustness against noisy sensor data. First, the initial fuzzy rules are extracted using a K-Nearest-Neighbor-based clustering approach. Next, a novel quasi-Levenberg-Marquardt (qLM) learning approach is proposed and applied to fine-tune the initial rules based on minimizing the cross-entropy loss function using a trust-region optimization method. Finally, complementary online learning is proposed to improve rules by encountering new labeled samples. The performance of the method is evaluated to classify patients and healthy subjects in different conditions, including the presence of noise or observing new samples. Moreover, the performance of the model is compared to some previous supervised and unsupervised machine learning approaches. The final Accuracy, Precision, Recall, and F1 Score of the proposed method are 97.61%, 97.58%, 99.02%, and 98.30%. [ABSTRACT FROM AUTHOR]

Wearable assistive robotics has emerged as a promising technology to supplement or replace motor functions and to retrain people recovering from an injury or living with reduced mobility. We developed delayed output feedback control for a wearable hip-assistive robot, the EX1, to provide gait assistance. Our purpose in this study was to investigate the effects of long-term exercise with EX1 on gait, physical function, and cardiopulmonary metabolic energy efficiency in elderly people. This study used parallel experimental (exercise with EX1) and control groups (exercise without EX1). A total of 60 community-dwelling elderly persons participated in 18 exercise intervention sessions during 6 weeks, and all participants were assessed at 5 time points: before exercise, after 9 exercise sessions, after 18 sessions, and 1 month and 3 months after the last session. The spatiotemporal gait parameters, kinematics, kinetics, and muscle strength of the trunk and lower extremities improved more after exercise with EX1 than in that without EX1. Furthermore, the effort of muscles over the trunk and lower extremities throughout the total gait cycle (100%) significantly decreased after exercise with EX1. The net metabolic energy costs during walking significantly improved, and functional assessment scores improved more in the experimental group than in the control group. Our findings provide evidence supporting the application of EX1 in physical activity and gait exercise is effective to improve age-related declines in gait, physical function, and cardiopulmonary metabolic efficiency among older adults. [ABSTRACT FROM AUTHOR]

Clinical gait analysis is a crucial step for identifying foot disorders and planning surgery. Automating this process is essential for efficiently assessing the substantial amount of gait data. In this study, we explored the potential of state-of-the-art machine learning (ML) and explainable artificial intelligence (XAI) algorithms to automate all various steps involved in gait analysis for six specific foot conditions. To address the complexity of gait data, we manually created new features, followed by recursive feature elimination using Support Vector Machines (SVM) and Random Forests (RF) to eliminate low-variance features. SVM, RF, K-nearest Neighbor (KNN), and Logistic Regression (LREGR) were compared for classification, with a Majority Voting (MV) model combining trained models. KNN and MV achieved mean balanced accuracy, recall, precision, and F1 score of 0.87. All models were interpreted using Local Interpretable Model-agnostic Explanation (LIME) method and the five most relevant features were identified for each foot condition. High success scores indicate a strong relationship between selected features and foot conditions, potentially indicating clinical relevance. The proposed ML pipeline, adaptable for other foot conditions, showcases its potential in aiding experts in foot condition identification and planning surgeries. [ABSTRACT FROM AUTHOR]

Mediolateral stability during walking can be controlled by adjustment of foot placement. Reactive activity of gluteus medius (GM) is modulated during the gait cycle. However, the mechanisms behind the modulation are yet unclear. We measured reactive GM activity and kinematics in response to a mediolateral platform translation during different phases of the gait cycle. Forward simulations of perturbed walking were used to evaluate the isolated effect of the perturbation and the GM response on gait stability. We showed that the potential of GM to adjust lateral foot placement and prevent collisions during swing varies during the gait cycle and explains the observed modulation. The observed increase in stance, swing or combined GM activity causes an outward foot placement and therefore compensates for the loss of stability caused by a perturbation early in the gait cycle. GM activity of the swing leg in response to a platform translation late in the gait cycle counteracts foot placement, but prevents collision of the swing foot with the stance leg. This study provides insights in the neuromechanics of reactive control of gait stability and proposes a novel method to distinguish between the effect of perturbation force and reactive muscle activity on gait stability. [ABSTRACT FROM AUTHOR]

The aim of this study was to evaluate the influence of balance on the spatiotemporal features, lower-limb kinematics, and center of mass (COM) of the non-faller elderly during walking. In this study, 20 healthy elderly women (age, 76.2 ± 5.6 years; height, 150.1 ± 3.2 cm; weight, 55.8 ± 9.0 kg) were enrolled. Based on the Berg balance scale (BBS), the elderly were classified into two groups: poor balance (PB; BBS scores <46; n = 10; 43.8 ± 1.8) and good balance (GB; BBS scores ≥46; n = 10; 50.4 ± 2.5). The two groups had no differences in terms of the spatiotemporal features and range of motion (ROM) of the vertical COM (all p > 0.05). The ROM of the mediolateral COM was greater in PB than in GB. Hip transversal movements in the two groups were different. The impairment of the lateral balance function might contribute to an increase in the incidence of fall events in the elderly with poor balance. [ABSTRACT FROM AUTHOR]

The aim of the study was to distinguish the hypokinetic gait disorder in idiopathic normal pressure hydrocephalus (NPH) patients from the gait decline in the elderly population by quantifying pathological gait parameter thresholds utilizing a multiple condition gait assessment. 55 NPH patients and 55 age-matched healthy subjects underwent a standardized gait assessment with eight gait conditions. Spatiotemporal gait parameters were assessed through a pressure-sensitive carpet. Statistical analysis consisted of a binary logistic regression (BLR) model, logistic curve-fit evaluated by a Chi-square goodness-of-fit-test, receiver operating characteristic models with area under the curves (AUC), and inverse BLR. Most discriminative gait parameter thresholds were observed in pace, gait cycle, and support gait domains. The most distinct gait conditions were preferred walking speed and semantic dual task. During preferred walking speed, the most significant gait parameter thresholds were stride length ≤ 1.02 m (sensitivity 0.93/specificity 0.91/AUC 0.96), gait velocity ≤ 0.83 m/s (0.80/0.91/0.93), double support phase ≥ 27.0% (0.96/0.76/0.91), and stride length coefficient of variation ≥ 3.4% (0.93/0.72/0.90). In conclusion, the hypokinetic gait disorder in NPH can be quantitatively differentiated from gait patterns of the elderly population. In future studies, this approach may be useful to differentiate clinical entities with similar gait disorders utilizing instrumented gait analysis procedures. [ABSTRACT FROM AUTHOR]

Robotic-assisted gait training (RAGT) devices allow intensive high repetition of the gait cycle in individuals with locomotor disability, with reduced therapist effort. In addition to usual rehabilitation, RAGT post-stroke improves the likelihood of regaining independent walking, with maximum efficacy identified in the acute and subacute phases of stroke. This study explores the usability and acceptance of RAGT among persons with stroke in an acute hospital setting and examines users' perceptions of two different modes of robotic assistance provided during rehabilitation. A mixed-methods approach comprised semi-structed interviews of end-user perspectives of RAGT in an acute hospital setting following stroke and two 10-point Likert scales rating how comfortable and how natural robotic gait felt using different assistance modes. Content analysis of qualitative data was undertaken with results synthesised by common meaning units. Quantitative data were reported using summary statistics, with Spearmann's correlation co-efficient examining the relationship between Likert scale ratings and measures of participants' stroke related disability. Ten individuals (6 men; 4 women; mean age of 64.5. ± 13 years) were recruited in an acute hospital setting following admission with a stroke diagnosis. Content analysis of interview transcripts identified discussion units centring around positive aspects of how helpful the device was, negative aspects related to set-up time, weight of the device and multiple instructions delivered during use. Initially participants identified that the device could look intimidating, and they feared falling in the device but they subsequently identified the correct mindset for using the device is to trust the technology and not be afraid. Mean ratings for device comfort (7.94 ± 1.4) and how natural walking felt (7.05 ± 1.9) were favourable. Interestingly, a strong relationship was identified, whereby the higher the level of disability, the more natural participants rated walking in the device during maximal assistance mode (rho = 0.62; p = 0.138). This study suggests individuals in the early phases of stroke perceive RAGT to be acceptable and helpful in the main, with some associated negative aspects. Walking in the device was rated as comfortable and natural. Those with greater disability rated the assisted walking as more natural. [ABSTRACT FROM AUTHOR]

Backward walking offers a unique challenge to balance and ambulation. This study investigated the characteristics of spatiotemporal gait factors and ankle kinematics during backward walking in people with chronic ankle instability. Sixteen subjects with chronic ankle instability and 16 able-bodied controls walked on a treadmill at their self-selected speed under backward and forward walking conditions. Gait speed, cadence, double limb support percentage, stride time variability, and three-dimensional ankle kinematics were compared between groups and conditions. During backward walking, both groups had significantly slower gait speed, lower cadence, and greater stride time variability. In addition, under backward walking condition, subjects in both groups demonstrated significant sagittal and frontal kinematic alternations, such as greater dorsiflexion and inversion following initial contact (0–27.7%, 0–25.0% of gait cycle respectively, p < 0.001). However, there were no significant differences between groups in any of the measured outcomes. This indicates that subjects with chronic ankle instability adapt to self-selected speed backward walking similarly to healthy controls. Assessments with more challenging tasks, such as backward walking with dual task and backward walking at fast speed, may be more appropriate for testing gait impairments related to chronic ankle instability. [ABSTRACT FROM AUTHOR]

Background and Objective: Crouch gait is the most common pathological gait pattern in cerebral palsy and is commonly seen in patients with spastic diplegia. It is characterized by excessive knee flexion throughout the stance phase of gait cycle. The aim of this review is to discuss the current literature about CG for a more comprehensive understanding. Methods: A literature review about various aspects of crouch gait in cerebral palsy was undertaken. This included its etiology and pathophysiology, biomechanics in crouch gait, natural history of untreated crouch gait, clinical and radiological evaluation and different modalities of available treatment. Results: The etiology is multifactorial and the pathophysiology is poorly understood. This makes its management challenging, thereby leading to a variety of available treatment modalities. Inadvertent lengthening of muscle–tendon units is an important cause and can be avoided. A meticulous clinical and radiological evaluation of patients, supplemented by observational and instrumented gait analysis is mandatory in choosing correct treatment modality and improving the treatment outcome. Younger children can be managed satisfactorily by various non-operative methods and spasticity reduction measures. However, crouch gait in cerebral palsy has a progressive natural history and surgical interventions are needed frequently. The current literature supports combination of various soft tissue and bony procedures as a part of single event multilevel surgery. Growth modulation in the form of anterior distal femur hemiepiphysiodesis for correction of fixed flexion deformity of knee has shown encouraging results and can be an alternative in younger children with sufficient growth remaining. Conclusions: In spite of extensive research in this field, the current understanding about crouch gait has many knowledge gaps. Further studies about the etiopathogenesis and biomechanics of crouch using instrumented gait analysis are suggested. Similarly, future research should focus on the long term outcomes of different treatment modalities through comparative trials. [ABSTRACT FROM AUTHOR]

Novel sensor technology enables new insights in the neuromechanics of human locomotion that were previously not possible. Here, we provide a dataset of high-density surface electromyography (HDsEMG) and high-resolution inertial measurement unit (IMU) signals, along with motion capture and force data for the lower limb of 10 healthy adults during multiple locomotion modes. The participants performed level-ground and slope walking, as well as stairs ascent/descent, side stepping gait, and stand-to-walk and sit-to-stand-to-walk, at multiple walking speeds. These data can be used for the development and validation of locomotion mode recognition and control algorithms for prosthetics, exoskeletons, and bipedal robots, and for motor control investigations. [ABSTRACT FROM AUTHOR]

This paper presents a method to regenerate lower limb joint angle trajectories during gait cycle by judging human intention using wearable sensor system. Myoelectric signals from user are used to detect the intention of gait initiation and gait phases. Multi-channel redundant fusion technique is implemented to obtain a robust stride time and gait phase calculation algorithm. Joint trajectories corresponding to particular gait events and phases are regenerated using a Radial basis neural network. The network is trained with joint angle data measured by Inertial Measurement Unit (IMU) from users with varying anthropomorphic features. Generated trajectory is adaptive to anthropomorphic as well as gait velocity variation. Contribution of this paper is in development of a wearable sensor system, multi-channel redundant fusion to calculate stride time and an adaptive gait trajectory generation algorithm. The proposed method of trajectory generation is used to regenerate lower limb joint motion in sagittal plane for wearable robotic devices like prosthesis and active lower limb exoskeleton. [ABSTRACT FROM AUTHOR]

Anterior cruciate ligament (ACL) deficient and reconstructed knees display altered biomechanics during gait. Identifying significant gait changes is important for understanding normal and ACL function and is typically performed by statistical approaches. This paper focuses on the development of an explainable machine learning (ML) empowered methodology to: (i) identify important gait kinematic, kinetic parameters and quantify their contribution in the diagnosis of ACL injury and (ii) investigate the differences in sagittal plane kinematics and kinetics of the gait cycle between ACL deficient, ACL reconstructed and healthy individuals. For this aim, an extensive experimental setup was designed in which three-dimensional ground reaction forces and sagittal plane kinematic as well as kinetic parameters were collected from 151 subjects. The effectiveness of the proposed methodology was evaluated using a comparative analysis with eight well-known classifiers. Support Vector Machines were proved to be the best performing model (accuracy of 94.95%) on a group of 21 selected biomechanical parameters. Neural Networks accomplished the second best performance (92.89%). A state-of-the-art explainability analysis based on SHapley Additive exPlanations (SHAP) and conventional statistical analysis were then employed to quantify the contribution of the input biomechanical parameters in the diagnosis of ACL injury. Features, that would have been neglected by the traditional statistical analysis, were identified as contributing parameters having significant impact on the ML model's output for ACL injury during gait. [ABSTRACT FROM AUTHOR]

After an initial ankle sprain, a relevant number of participants develop chronic ankle instability (CAI). Compensatory strategies in patients with CAI may change the inter-limb symmetry needed for absorbing movement-related forces. Accordingly, an increased risk of injury can occur. The present study aimed to compare the inter-limb asymmetry of kinetic and electromyography between individuals with CAI and without a history of an ankle sprain (Non-CAI) during walking. In this cross-sectional study, fifty-six athletes (28 CAI; 28 Non-CAI) participated. Participants walked at a comfortable pace over level ground while vertical ground reaction force (vGRF) and muscle activity of the tibialis anterior, peroneus longus, medial gastrocnemius, and gluteus medius were recorded. Inter-limb asymmetry during walking was calculated for each of the variables. Patients with CAI exhibited a greater inter-limb asymmetry of the first peak of vGRF, time to peak vGRF, and loading rate (P < 0.001), as well as presenting a greater inter-limb asymmetry of peroneus longus activity (contact phase) (P = 0.003) and gluteus medius activity (midstance/propulsion phase) (P = 0.010) compared to the Non-CAI group. No other differences in vGRF or muscles activity were observed between the groups. Our findings indicate that patients with CAI walk with greater inter-limb asymmetry in vGRF and muscle activity in different phases of the gait cycle compared to Non-CAI group. Our results could inform future studies on gait training aimed to reduce asymmetry during walking in patients with CAI. [ABSTRACT FROM AUTHOR]

The gait recognition is to recognize an individual based on the characteristics extracted from the gait image sequence. There are many researches for the gait recognition which use diverse kinds of information such as shape of gait silhouette, motion variation caused by walking, and so on. In general, shape information is more useful for recognition. However, shape information is influenced by a variety of factors, which degrade the recognition performance. Moreover, the information used in most of those studies might be able to be extracted after all of one or more sequences of the gait cycle are known. And it is also hard to discriminate the gait cycle from given gait sequences exactly by the online approach. In regard to these difficulties, we propose a novel gait recognition method based on the multilinear tensor analysis. To recognize the cyclic characteristic of gait without an exact division for the gait cycle, this paper's propose is the method to form the accumulated silhouette and then describes those as the tensor. For the accumulated silhouette proposed by this paper, the image sequence of one gait cycle is divided into four sections in the training phase. However, discrimination for the gait cycle in the training phase is not directly related to the recognition phase, thus the online approach is possible. We first form the accumulated silhouettes for every individual using gait silhouettes within each section. And then, we represent these accumulated silhouettes as the tensor. Using a multilinear tensor analysis, we compute the core tensor which governs the interaction between factors organizing the original tensor, and then compose the basis to recognize the individual in the online recognition framework. Finally, we recognize the individual using the computation of similarity based on the Euclidean distance, which is more suitable to our method. We verify the superiority of the proposed approach via experiments with real gait sequences. [ABSTRACT FROM AUTHOR]

Frailty is associated with gait variability in several quantitative parameters, including high stride time variability. However, the associations between joint kinematics during walking and increased gait variability with frailty remain unclear. In the current study, principal component analysis was used to identify the key joint kinematics characteristics of gait related to frailty. We analyzed whole kinematic waveforms during the entire gait cycle obtained from the pelvis and lower limb joint angle in 30 older women (frail/prefrail: 15 participants; non-frail: 15 participants). Principal component analysis was conducted using a 60 × 1224 input matrix constructed from participants' time-normalized pelvic and lower-limb-joint angles along three axes (each leg of 30 participants, 51 time points, four angles, three axes, and two variables). Statistical analyses revealed that only principal component vectors 6 and 9 were related to frailty. Recombining the joint kinematics corresponding to these principal component vectors revealed that frail older women tended to exhibit greater variability of knee- and ankle-joint angles in the sagittal plane while walking compared with non-frail older women. We concluded that greater variability of knee- and ankle-joint angles in the sagittal plane are joint kinematic characteristics of gait related to frailty. [ABSTRACT FROM AUTHOR]