Search Results (709)

Three-dimensional instrumented gait analysis is widely used to guide surgical decision-making in children with cerebral palsy (CP), but its interpretation is time-consuming and prone to inter-rater variability. In this single-centre pilot study, we investigated whether a generative large language model (LLM) could consistently generate gait deviation findings and surgical procedure suggestions that align with expert judgement. Kinematic features for lower-limb joints across the gait cycle, stance, and swing were extracted from eight children with unilateral CP using the open-source GaitSharing Toolkit and a structured prompt, then submitted three times per patient to OpenAI’s GPT-5.5 model. The model assessed 28 kinematic deviations and 12 surgical procedure groups using majority voting. One gait analyst and two paediatric orthopaedic surgeons independently rated outputs on a 0–2 ordinal scale, blinded to all clinical information beyond the kinematic curves and diagnosis. Agreement was summarised descriptively as the percentage of the maximum attainable score with 95% confidence intervals (CIs), and quadratic-weighted Cohen’s kappa was used to quantify inter-surgeon agreement. Agreement with the gait expert was highest at the hip (90.6%) and lowest at the knee, particularly in the transverse plane (65.2%). For surgical procedures, agreement with the LLM reached 83.9% and 73.4% for the two surgeons, with the tibialis anterior procedure showing the lowest concordance. Inter-surgeon agreement was 79.2% (95% CI 71.9–85.4) with a kappa of 0.59 (0.47–0.70), indicating moderate agreement. The LLM showed high self-consistency (>90% across runs). These preliminary findings suggest that generative LLMs may be feasible as assistive tools in clinical gait analysis for deviation detection and future treatment planning and should be interpreted as hypothesis-generating, warranting confirmation in larger, more diverse cohorts. Full article

Introduction: Chronic stroke frequently causes structural and functional impairments in the tibialis anterior (TA) muscle, leading to foot-drop and altered gait kinematics. We investigated the relationship between ultrasonographic parameters of the TA muscle and functional gait performance in chronic stroke survivors. Methods: This cross-sectional, observational pilot study evaluated eight consecutive chronic stroke patients (>6 months post-stroke). Structural parameters of TA (rest and contraction thickness, pennation angle) were documented by ultrasound imaging. A dynamic feature calculated was the Contraction Index (CI = effort/resting thickness). Functional metrics included the Medical Research Council (MRC) muscle strength scale and the 10-Meter Walk Test (10MWT) for gait velocity. Results: Statistical analysis revealed a moderate negative correlation trend between resting TA muscle thickness and functional gait speed (r = −0.618, p = 0.102). Conversely, a moderate positive correlation trend was found (r = 0.548, p = 0.160). Ultrasound imaging successfully differentiated three distinct pathological phenotypes: an atrophic phenotype (low pennation angle, flaccid muscle failure), a severely shortened spastic phenotype (increased resting thickness, high pennation angle, pathological CI < 1.0), and a spastic co-contraction loop. Patients with the atrophic phenotype achieved high mechanical efficiency with an ankle–foot orthosis (AFO), whereas the spastic phenotype exhibited resistance against the orthotic device. Conclusions: Musculoskeletal ultrasound provides objective parameters for post-stroke TA muscle remodeling and contributes to completing the assessment and therapy. Identifying specific structural muscle phenotypes allows clinicians to optimize target-specific neurorehabilitation strategies, predict AFO efficiency, and guide antispastic interventions such as botulinum toxin injections. Full article

Gait symmetry analysis has emerged as an important tool in rehabilitation engineering and neurological disorder assessment, as it provides clinically relevant indicators of mobility impairment and gait abnormalities. The proposed framework integrates gait symmetry variability, statistical gait features and Dynamic Time Warping (DTW)-based temporal alignment to enhance robustness against gait variations and irregular walking patterns. A hybrid feature vector comprising DTW similarity scores, the hybrid gait symmetry index (GSI), and statistical gait descriptors was employed to classify gait patterns into five categories: normal, slow, medium, fast, and abnormal. The system was implemented as a wearable edge-computing platform using an NI myRIO device equipped with a tri-axial Inertial Measurement Unit (IMU) mounted on the subject’s body. The onboard FPGA performs real-time signal preprocessing, GSI computation, feature extraction, constrained DTW matching, and gait classification using fixed-point streaming architectures and BRAM-based buffering. Meanwhile, the embedded ARM processor manages TCP/IP communication and transmits real-time gait information to a remote monitoring workstation via a WiFi interface for visualization and analysis. Operating at a clock frequency of 100 MHz, the complete architecture achieves an end-to-end processing latency of approximately 4 ms. The proposed FPGA-based implementation provides low-latency, energy-efficient, and real-time gait analysis, making it well suited for wearable rehabilitation systems, assistive healthcare devices, and continuous mobility monitoring applications. Full article

Background/Objectives: Commercial wearables are increasingly used after total hip and knee arthroplasty (THA/TKA) to record activity and gait outside the clinic and may help tailor recovery pathways. We reviewed the evidence on commercial devices, digital phenotyping, and data-informed rehabilitation after arthroplasty. Methods: We performed a structured narrative search of PubMed and Google Scholar. Of 916 records, 63 duplicates were removed, 853 were screened, 245 full texts were assessed, and 76 publications were included. Findings were synthesized thematically; no formal risk-of-bias assessment or evidence grading was undertaken. Results: Recovery differed by outcome and procedure. PROMs commonly improved within 1–3 months, while gait-quality measures recovered over about 7–13 weeks after THA and 13–24 weeks after TKA. Frequency-domain gait features, sample entropy, and machine-learning models provided information beyond step counts. Tracker-based interventions most consistently increased postoperative steps when introduced early and combined with behavior-change support, while digital rehabilitation was generally non-inferior to conventional rehabilitation. Wrist-worn consumer devices remained inaccurate with gait aids, compliance definitions varied, and higher step counts did not consistently correspond to better PROMs. Conclusions: Commercial wearables already support phenotyping and remote follow-up of individual recovery trajectories, but evidence for fully adaptive rehabilitation remains limited. Full article

Lower-limb amputation is associated with persistent gait asymmetries and functional limitations that are not fully captured by conventional clinical outcome measures. This study aimed to develop and prospectively validate a wearable sensor-based Gait Goodness Score (GGS) derived from ensemble classifiers to summarize overall gait quality during supervised clinical walking. The algorithm was previously trained using inertial measurement unit data and clinically meaningful temporal–spatial features. In the present prospective, observational validation study, medically stable adults with lower-limb amputation performed supervised 10 m walk tests in outpatient rehabilitation settings, during which step-based GGS values were collected. Associations between GGS and established clinical measures, including walking velocity, Amputee Mobility Predictor (AMP) score, and Timed Up and Go (TUG) durations, were examined. GGS demonstrated significant differences across functional levels and amputation levels and showed strong associations with walking velocity and AMP score, with a significant moderate-to-fair association also observed for TUG and PLUS-M. These findings support the validity of GGS as a quantitative, sensor-derived metric of gait quality during supervised clinical walking in individuals with lower-limb amputation. Full article

Parkinson’s disease (PD) is associated with gait impairment, bilateral asymmetry, and increased gait variability, highlighting the need for objective and interpretable wearable gait assessment. Plantar insole recordings directly capture foot–ground loading, but their use in PD assessment is often limited by global or low-order descriptors that do not fully represent regional loading organization. This study proposes a region-specific information-theoretic framework for PD gait assessment using wearable plantar-pressure insoles. Bilateral plantar insole signals were reorganized into five anatomical regions: heel, rearfoot, midfoot, forefoot, and toe. Self-information index (SII), Shannon entropy (EN), negentropy (NEG), sample entropy (SEN), and Kullback–Leibler divergence (KL) features were extracted to characterize self-information fluctuation, probabilistic uncertainty, non-Gaussian organization, temporal irregularity, and directional distributional discrepancy in plantar-pressure dynamics. The resulting feature representation was evaluated at gait-cycle, walking-recording, and subject-independent levels using conventional classifiers, ablation analysis, subject-balanced cycle aggregation, and an information-theoretic three-dimensional feature-space rule model (ITFS-RM). KNN achieved an accuracy of 0.9668 at the gait-cycle level, and MLP achieved an accuracy of 0.9344 at the walking-recording level. Under stricter subject-independent evaluation, the accuracy was 0.8475, and subject-balanced-cycle aggregation achieved an accuracy of 0.8655. Region-specific analysis and ablation experiments showed spatially heterogeneous HC–PD differences, with the toe region showing the most consistent contribution. SII, KL, and NEG provided stable discriminative contributions, particularly in toe-related and regional-transition features. ITFS-RM provided explicit feature combinations, value ranges, and spatial rule boundaries for interpretable walking-recording level and subject-grouped separation. These results support region-specific information-theoretic analysis as an interpretable representation of plantar-pressure dynamics for PD gait assessment and emphasize the need for subject-wise validation when repeated walking recordings are available. Full article

Background: The spinocerebellar ataxias (SCA) are a group of multiple inherited disorders. The molecular diagnosis of the specific type of SCA is challenging due to the diversity of hereditary patterns and genetic anomalies. Certain key clinical features can help shorten the differential diagnosis. Cases: At our center, we evaluated three patients presenting with gait disturbances. They were assessed by multiple hospital departments, and, due to suspected spinocerebellar ataxia, comprehensive genetic and ophthalmologic testing was performed, leading to diagnoses of SCA2, SCA7, and SCAR32. These cases highlight the importance of multidisciplinary evaluation for accurate diagnosis of complex neurogenetic disorders. All patients experienced prolonged diagnostic delays, requiring years of multiple consultations across different hospitals before reaching a definitive diagnosis. Literature Review: Among all published reports of patients with SCA, we consider that five distinct key clinical features can categorize many patients suffering from a possible SCA into different groups and reduce the potential differential diagnosis. Only two types of SCA presented macular anomalies, ten optic nerve involvements, one erythrokeratodermia, four hearing losses, and four dementias. Conclusions: Early referral to a multidisciplinary evaluation might help narrow the differential diagnosis of specific types of SCA, guiding the diagnostic work-up toward SCA1 and SCA7 in patients with macular involvement, or toward SCA1, SCA2, SCA3, SCA7, SCAR3, SCAR9, SCAR21, SCAR29, SCAR31, and SCAX3 in those with optic nerve involvement. Full article

Walking becomes less economical with age, but the links between detailed gait mechanics and metabolic energy expenditure remain unclear. This study examined multidimensional associations between gait characteristics and walking energetics in community-dwelling older adults. Eighty-five participants, including 62.4% women, completed a steady-state walking task while metabolic energy expenditure was measured using portable indirect calorimetry. Four energetic outcomes were analyzed: walking cost as a percent of peak metabolic capability, steady-state walking metabolic rate, gross cost of transport, and reserve cost of transport. Gait was assessed using an instrumented walkway and summarized across spatial, temporal, and variability domains, yielding 28 gait parameters. Pearson correlations were calculated between each gait parameter and each metabolic outcome, with significance set at p < 0.05. Across 28 gait features and four metabolic outcomes, clear association patterns emerged. Spatial parameters showed the strongest and most consistent relationships with metabolic cost, suggesting that step length, stride length, and related forward-progression measures are closely tied to energetic demand during steady-state walking. Temporal parameters showed meaningful but generally weaker associations, while variability-based metrics also demonstrated moderate significant correlations. These findings provide a quantitative framework linking gait mechanics with walking energetics and may help identify mobility biomarkers and intervention targets in older adults. Full article

Sit-to-walk (STW) is a short-duration, high-load, multijoint transition requiring rapid lower-limb neuromuscular coordination across seat-off, load transfer, and gait initiation. Surface electromyography (sEMG)-based prediction of knee and ankle joint moments may support motor function evaluation and inform future assistive-control applications, but existing models remain limited in modeling cross-muscle sEMG feature interactions and mitigating phase-dependent prediction errors. This study developed an anatomy-informed framework for sEMG-driven moment prediction during STW. The model encoded sEMG channels from thigh and shank muscles into separate anatomical branches. Cross-Attention was used to model task-relevant intersegmental interactions, and BiLSTM was applied to capture short-term temporal dependencies. Eighteen healthy participants performed STW trials while sEMG, three-dimensional kinematics, and ground reaction forces were synchronously collected. Knee and ankle moments were calculated using inverse dynamics and used as reference targets. Among six models, the Cross-Attention model achieved the lowest test-set overall error, with an Overall nRMSE Fixed of 4.51%; the knee peak error in the P3 unloading phase was 16.17%. Ablation experiments indicated that Cross-Attention, BiLSTM temporal modeling, anatomical branch separation, and joint-specific output mapping contributed to prediction performance. This framework provides an interpretable approach for sEMG-driven multijoint moment prediction in complex non-stationary movements. Full article

Background and Objectives: Accelerated rehabilitation after anterior cruciate ligament reconstruction (ACLR) is widely implemented, yet its definition and distinguishing characteristics remain inconsistently described in the literature. This scoping review examined how accelerated rehabilitation after ACLR is defined, described common protocol features, and identified elements distinguishing it from conventional rehabilitation. Materials and Methods: A scoping review was conducted using systematic searches of Medline (PubMed), Embase, and Web of Science from 1 April 1967 to 26 October 2025. Studies including patients aged 16 years or older who underwent primary ACLR that reported any form of accelerated rehabilitation or early progression relative to conventional protocols were eligible for inclusion. Results: Of 6002 screened records, 64 studies met the inclusion criteria. Accelerated rehabilitation was consistently characterized by early restoration of knee range of motion, early full weight-bearing, rapid gait normalization, early initiation of closed and open kinetic chain exercises, and avoidance of prolonged immobilization. However, definitions varied substantially across studies. Substantial heterogeneity was observed in progression timelines, bracing and crutch use, and return-to-sport criteria. Conclusions: Accelerated rehabilitation after ACLR appears to represent a brace-free, criterion-based, function-oriented approach emphasizing early restoration of knee extension, progressive loading, and individualized progression rather than simply shortened timelines. Establishing consensus definitions and standardized reporting is necessary to improve comparability across studies and facilitate translation into clinical practice. Full article

Accurate identification of gait abnormalities is fundamental to clinical evaluation in both veterinary medicine and translational research. Commonly used descriptors such as ataxia, paresis, hypermetria, and lameness are frequently applied as single-word summaries of complex movement disorders, yet these terms lack precise, universally accepted operational definitions. Dictionary definitions are often generic or ambiguous, leading to variability in interpretation among clinicians, students, and researchers. This review evaluates commonly used neurologic and gait-related terminology in dogs, examining historical origins, denotation and connotation, as well as the limitations of current usage in a clinical setting. For each term, clinically observable features are delineated to establish functional, operational definitions grounded in movement analysis and neurophysiologic processes. The goal of this reappraisal is to enhance clarity, consistency, and diagnostic accuracy in veterinary neurology by establishing standardized descriptors for gait abnormalities in clinical practice, teaching, and research. Full article

Repeated sit-to-stand and stand-to-sit transitions load the knee extensors and may contribute to work-related musculoskeletal disorders. Reducing this load requires assistive devices and monitoring of knee function, which depend on real-time onset/offset detection and direction-aware classification of each transition. However, no prior wearable surface electromyographic system has delivered this capability for real-time. This study presents a deep learning method that computes both onset/offset detection and direction discrimination of sit-to-stand and stand-to-sit in a developed wearable surface electromyographic system in real-time. Two ESP32-S3 nodes and a hub record from the vastus lateralis and vastus medialis and run a per-burst convolutional detector, while the hub runs a dual-branch classifier with seventeen handcrafted features. Trained offline on the public Gait120 dataset, the networks are deployed unchanged with embedded-firmware parity to the MATLAB reference. Under leave-one-subject-out evaluation on Gait120, the offline classifier separated each transition with 99.6% accuracy and the detector achieved 96.6% completeness. In real-time recordings from thirty healthy adults, the system retained 85.6% classification and 82.0% detection accuracy, with ≈100 ms latency and a 618 KB network footprint. Results show that a low-power wearable delivers combined detection and phase discrimination in real-time, supporting the potential application in assistive-device control and knee-joint monitoring. Full article

Concussion is associated with persistent motor control deficits that may not be detected using standard clinical assessments. This study examined differences in average velocity, average step length, and single- and double-support percentages during gait under increasing task demands in individuals with a history of concussion. Sixty participants aged 18 to 35 years were recruited, including 32 individuals with a concussion within the past 3 to 18 months and 28 healthy controls. Gait data were collected using an instrumented pressure-sensitive walkway across four conditions: single-task and dual-task walking, with and without obstacles. Repeated-measures analyses of covariance were used to assess group and condition effects, with sex as a covariate, which showed a significant group-by-condition effect for step length, single-support percentage, and double-support percentage. These findings identify step length, single-support percentage, and double-support percentage as candidate sensor-derived gait biomarkers for detecting persistent post-concussion motor control alterations. The results directly support the use of pressure-based gait sensing to quantify deficits missed by conventional clinical measures; however, future work is needed to determine whether these features translate to wearable or real-world monitoring systems. Full article

Gait analysis is used in the diagnosis, rehabilitation, and longitudinal monitoring of degenerative spinal disease (DSD). However, conventional gait assessment commonly depends on subjective visual observation or laboratory-based motion-capture systems, which restrict accessibility and routine clinical use. This study presents an edge-enabled real-time gait analysis framework for DSD using two ankle-worn inertial measurement units (IMUs). The proposed framework integrates causal gait-event detection with spatiotemporal gait estimation, including stride length, stride height, stride frequency, and swing ratio, across Regular-, Toe-, Heel-, and Tandem-Walk tasks. To improve the stability of wearable gait estimation, the framework incorporates a cycle-wise initial sensor-orientation correction strategy with inter-cycle horizontal velocity continuity, reducing reliance on conventional zero-velocity update (ZUPT) resetting. A percentile-referenced, separability-weighted composite score was also developed to combine average gait performance, step-to-step variability, and gait asymmetry into an interpretable clinical index. Algorithm validation was conducted using an optical motion-capture system as the reference standard. The proposed framework, however, is intended for deployment using ankle-worn IMUs and edge-based computation without requiring optical cameras, reflective markers, or dedicated motion-capture laboratories during routine operation. Experimental results showed centimeter-level spatial estimation accuracy. The composite scoring framework achieved accuracy/F1-scores of 0.971/0.981 for Regular-Walk, 0.934/0.956 for Toe-Walk, 0.955/0.970 for Heel-Walk, and 0.876/0.919 for Tandem-Walk. Feature analysis indicated that stride length, stride frequency, swing ratio, and step-to-step variability provided the greatest discrimination between healthy controls and spinal patients, with stronger group separation observed in Regular-, Toe-, and Heel-Walk tasks. These results suggest that ankle-mounted IMU sensing combined with lightweight edge-based computation and interpretable gait scoring may provide a practical approach for point-of-care gait assessment and remote functional monitoring in DSD. The proposed system operates using only two ankle-mounted IMUs during routine deployment, while optical motion capture was employed exclusively as a laboratory reference for validation. Full article

Hoof slipping in dairy cows is a subtle, transient hoof motion event distinct from lameness or falling, with short duration, limited displacement, and close resemblance to normal gait, making automated detection particularly challenging; relevant methods remain scarce. This study proposes a cascaded detection framework based on improved DeepLabCut and NeuFlow v2 for automated hoof-slipping detection and distance estimation in Holstein dairy cows. The four-stage framework covers hoof key point localization, pixel-level optical flow fusion, motion parameter curve feature extraction, and Random Forest classification. The framework was developed on Dataset 1, which contained 115 single-cow side-view videos. Of these, 31 contained slipping events and 84 were normal walking. It was further assessed on a smaller second-farm dataset of 17 single-cow videos (Dataset 2). ResNet-50 with a Coordinate Attention mechanism was adopted as the backbone, reducing mean four-hoof localization RMSE to 2.80 pixels across five independent training runs, showing a 15.2% improvement over the baseline, and outperforming YOLOv8s-Pose. NeuFlow v2 was applied to extract the localized optical flow from hoof regions, yielding velocity and directional curves from which slipping features were derived. The Random Forest classifier achieved an accuracy of 98.9%, precision of 93.3%, recall of 90.3%, F1 score of 91.8%, and AUC of 0.995, outperforming MViT, SlowFast, and STME. The slipping distance estimation RMSE was 1.22 pixels. With the localisation model retrained on new farm frames, the method reached comparable performance on the second farm, suggesting preliminary cross-farm generalisability that warrants larger-scale validation. The proposed framework provides a non-invasive basis for early hoof-health monitoring and welfare-oriented farm management. Full article