Search Results (1,244)

Background and Clinical Significance: Wernicke encephalopathy (WE) is a potentially reversible neurological emergency caused by thiamine deficiency. Although classically associated with alcohol use, it may also occur after gastrointestinal surgery and prolonged nutritional compromise, where diagnosis can be obscured by postoperative delirium or metabolic encephalopathy. Case Presentation: A 58-year-old woman with no history of alcohol use underwent a Whipple procedure for pancreatic head adenocarcinoma. Her postoperative course included an anastomotic leak, prolonged open-abdomen management, severely restricted oral intake, and long-term total parenteral nutrition. Approximately 1.5 months after surgery, she developed fluctuating confusion, visual hallucinations, marked recent-memory impairment, diplopia, nystagmus, and gait ataxia, initially interpreted as postoperative delirium/metabolic encephalopathy. Approximately two years later, persistent anterograde amnesia, cerebellar and oculomotor findings, and neuropsychological deficits in attention, memory, and executive functions led to clinical suspicion of previous WE. Re-evaluation of the acute MRI showed bilateral medial thalamic and periaqueductal FLAIR hyperintensities that were absent on follow-up imaging, supporting the retrospective diagnosis; the persistent amnestic syndrome was compatible with Korsakoff syndrome. B-complex vitamin supplementation was initiated, followed by 10 weekly 60 min cognitive rehabilitation sessions and a family-supported home program. Family members reported improved retention of recent information and greater daily independence, but formal post-treatment neuropsychological reassessment could not be completed. Conclusions: This case highlights the difficulty of recognizing non-alcoholic WE during a complicated postoperative course. The observed functional changes cannot be attributed specifically to rehabilitation because vitamin supplementation, nutritional recovery, spontaneous recovery, and rehabilitation occurred within the same follow-up period. Full article

This paper presents a wearable soft strain sensor based on a commercially available conductive thermoplastic polyurethane (cTPU) 3D-printed as an auxetic soft metamaterial for human activity recognition. The growing demand for flexible and wearable electronics, driven by advances in artificial intelligence, highlights the importance of such sensors in healthcare, medical rehabilitation, soft robotics, and human–machine interfaces. The auxetic cTPU sensor was mechanically and electrically characterized through empirical measurements and validated against numerical simulations. A single sensor mounted on a knee brace was used to collect gait signals across four activities: running, walking, standing, and sitting. Two classification approaches were investigated. A Long Short-Term Memory (LSTM) network was trained directly on the raw time-series signal, with the best configuration achieving 96% accuracy using Relative Standard Deviation Normalization with 50 hidden units. Traditional machine learning models, namely Random Forest and XGBoost, were trained on 30 extracted time-domain and frequency-domain features per motion cycle, achieving 100% and 97.33% accuracy, respectively, under five-fold cross-validation. To enhance model transparency, explainability analysis using SHAP identified power spectral density and the first harmonic frequency as the most consistently influential features across both models, with dynamic activities driven by frequency characteristics and stationary activities distinguished by signal mean amplitude. The results demonstrate that auxetic cTPU soft strain sensors combined with machine learning and explainable artificial intelligence provide an accurate and interpretable solution for wearable human activity recognition, highlighting their potential for applications in robotics, healthcare, and human–robot interfaces. Full article

Background: The current evidence regarding the impact of augmented reality (AR) rehabilitation on motor function in children with cerebral palsy (CP) remains limited. Objectives: This study aimed to evaluate whether integrating AR motor training (ARMT) into conventional rehabilitation enhances gait and balance outcomes compared to conventional rehabilitation alone. Methods: Forty children aged 7 to 12 years with unilateral or bilateral spastic CP were randomly assigned to receive either ARMT integrated into a 4-week conventional rehabilitation program (n = 20) or conventional rehabilitation alone (n = 20). The ARMT replaced 20–30 min of standard motor therapy five days per week. Gait and balance were assessed pre- and post-intervention using an instrumented 10 m walk test with the G-Walk sensor, the MABC-2 one-leg balance task, and the Pediatric Balance Scale. Due to unavailability of data from five control-group participants, analyses were conducted on 35 children using available-case mixed-model approaches. Results: Nominally significant between-group differences in pre- to post-intervention changes were observed for left stride cycle duration (mean difference: −0.083 s; 95% confidence interval [CI]: [−0.166, −0.0002]; p = 0.049), right stride cycle duration (−0.095 s; 95% CI: [−0.185, −0.006]; p = 0.038), the coefficient of variation of the first double-support phase on the right side (−8.91 percentage points; 95% CI: [−17.39, −0.44]; p = 0.040), and the left propulsion index (2.42 m/s²; 95% CI: [0.39, 4.45]; p = 0.021). No statistically significant differences were detected between groups for balance measures, gait speed, stride length, gait quality, or symmetry. Conclusions: Incorporation of four weeks of ARMT into conventional rehabilitation did not yield substantial additional benefits for gait or balance in this cohort. The observed effects on gait timing, variability, and propulsion are exploratory and warrant validation in larger, adequately powered studies with pre-specified primary outcomes. Full article

Human Activity Recognition (HAR) plays a significant role in various applications, from learning a discipline to physical rehabilitation. In older adults, activity patterns can indicate levels of frailty, which helps inform the design of physical training programs to prevent falls and maintain mobility. HAR sensing ranges from wearable sensors such as IMUs and RGB cameras to video, specialized gait laboratories, perturbation units, VR, and other modalities. This paper presents a comprehensive study of depth-based, skeleton-driven HAR and age group recognition (AGR) using data collected from real-world nursing home environments. Depth sensors offer a privacy-preserving and non-invasive alternative to wearable and RGB-based systems, enabling continuous 24-h monitoring without requiring user compliance. We systematically evaluate multiple modeling paradigms, including classical machine learning models (DT, RF, KNN, SVM, HMM, HMM+SVM), sequence-based models (LSTM, TCN, ARNN), and graph-based approaches, using skeletal joint data extracted from depth images. Experiments are conducted on two heterogeneous datasets: NTU RGB+D (younger adults) and ETAP-DID (older adults). We analyze the impact of different joint subset configurations (full-body, limb-only, leg-only, and torso-only) and compare raw joint representations with handcrafted time-series features (TSFEL) for frame-based HAR. Beyond activity recognition, we introduce an AGR pipeline to distinguish younger from older adults based on skeletal motion patterns. We investigate multiple feature representations, including absolute joint positions, root-relative coordinates, bone vectors, and joint velocities, and provide interpretability through feature importance and saliency analysis to identify age-discriminative joints and motion cues. Our study provides a comprehensive analysis of various HAR models applied to depth data, examining model performance and the contribution of joint-based features to HAR and AGR. Our study highlights the potential for personalized privacy-preserved monitoring and intervention in nursing homes. Full article

Wearable technologies are reshaping gait rehabilitation by shifting assessment and therapy from intermittent, clinic-based observation to continuous, data-driven, and adaptive care. This narrative review synthesizes the biomechanical basis of gait instability, the architecture and classification of wearable rehabilitation systems, and the clinical evidence for motion sensors, smart insoles, biofeedback devices, robotic and orthotic wearables, neuromodulatory systems, immersive platforms, and artificial intelligence (AI)-enabled closed-loop interventions. Its main contribution is an integrated framework that links AI, digital biomarkers, device components, adaptive control, and translational implementation, rather than treating wearable rehabilitation as a device-only or disease-specific topic. Current evidence indicates that these technologies can improve gait speed, symmetry, balance, endurance, fall-risk monitoring, and dual-task performance in neurological, musculoskeletal, frailty-related, and aging populations. However, the field is still limited by heterogeneous protocols, small samples, limited longitudinal validation, insufficient device standardization, usability barriers, cybersecurity concerns, uncertain reimbursement, and restricted interoperability with healthcare systems. Future progress will depend on multimodal sensor fusion, explainable and federated AI, digital twins, adaptive wearable robotics, tele-rehabilitation pathways, and large-scale pragmatic trials that validate effectiveness in real-world rehabilitation settings. Full article

Balance impairment threatens mobility and independence in older adults with cardiovascular disease, yet cardiac rehabilitation (CR) has traditionally prioritised aerobic capacity and cardiovascular outcomes. This narrative review examines the mechanisms and assessment of balance impairment, balance recovery within CR, and implications for clinical practice. Balance impairment reflects interactions among musculoskeletal, sensory, cognitive–motor, and cardiovascular constraints that affect different domains of postural control to varying extents. Standardised clinical measures do not capture these domains equally, and single scores or completion times can obscure the deficits underlying poor performance. Instrumented and wearable technologies extend clinical assessment by quantifying postural sway and gait, helping to distinguish broader mobility gains from recovery within specific balance domains. To translate this distinction into practice, this review recommends individualised, balance-focused CR, with assessment guiding task-specific training and virtual reality or exergaming providing graded practice and performance feedback to promote mobility and independence. Full article

Pathological gait classification plays an important role in objective motor function assessment, early clinical screening, and rehabilitation evaluation. However, traditional clinical gait assessment methods are highly dependent on expert experience and may fail to detect subtle gait abnormalities. Moreover, existing inertial measurement unit (IMU)-based gait recognition methods often rely on single-sensor configurations or single-scale temporal models, limiting their ability to capture complex pathological gait patterns. In this study, a convolutional neural network–bidirectional long short-term memory–temporal convolutional network (CNN-BiLSTM-TCN) multi-branch feature fusion framework was proposed for pathological gait classification using a publicly available clinical multi-inertial measurement unit dataset containing 260 subjects. The proposed model employs three parallel branches to extract local instantaneous motion variations, continuous temporal dynamics, and relatively broader temporal dependencies within the 2 s input window, respectively, followed by feature-level fusion and end-to-end joint optimization. Experimental results show that the proposed model achieves a test accuracy of 0.9818 and an F1-score of 0.9700, outperforming conventional machine learning methods, single-branch models, voting-based fusion methods, and other temporal models, including Support Vector Machine (SVM), Temporal Convolutional Network (TCN), and Convolutional Neural Network-long short-term memory (CNN-LSTM). Five repeated experiments with stratified random splits demonstrate minimal performance variation, indicating good robustness and stability. The proposed framework provides a potential approach for pathological gait screening and quantitative rehabilitation assessment. Full article

Portable electromyography biofeedback (EMG-BFB) may support gait rehabilitation, but whole-gait responses across increasing portable auditory feedback goals are unclear. Twenty-four adults completed baseline treadmill walking and four counterbalanced right medial gastrocnemius activation-goal conditions set at 20%, 40%, 60%, and 80% above baseline; 23 contributed primary biomechanical data. Treadmill speed was adjusted within each condition to support achievement of the activation goal. Outcomes included walking speed, ground-reaction forces, force-time metrics, step length, temporal measures, and asymmetry. Repeated-measures MANOVA, outcome-specific repeated-measures ANOVAs, dose-response coefficients, bootstrap intervals, and leave-one-participant-out analyses were used. The combined gait-biomechanics outcomes differed significantly across activation-goal conditions (p < 0.001), with large condition effects for walking speed, propulsion, braking magnitude, and step length. From baseline to the highest goal, treadmill speed increased from 1.07 to 1.44 m/s, mean propulsion by 0.086 N/BW, braking magnitude by 0.109 N/BW, and mean step length by 0.150 m. Exploratory speed-adjusted models retained anterior–posterior and vertical loading-response associations but not peak propulsion; activation goal and achieved speed were strongly collinear. Unilateral feedback was not associated with systematic step-length or step-time asymmetry. Portable auditory EMG-BFB at increasing activation goals was accompanied by coordinated changes across gait mechanics. Full article

Lower limb mobility impairments resulting from neurological diseases, trauma injuries or aging significantly impact the quality of life of individuals by limiting their autonomy. Rehabilitation plays an important role in addressing the challenges of these impairments, with early detection of changes in the functional state of individuals being essential. This enables therapies to be adjusted based on the current condition of the patient, thereby enhancing their effectiveness. Such early detection, however, requires continuous assessment by specialists, which is unfeasible given the existing limited resources. Since gait is a reflection of the physical and mental states of each individual, its continuous monitoring and subsequent data analysis can serve as a valuable tool for the aforementioned objective. This study proposes a methodology that, based on continuous gait monitoring data, detects significant changes in the functional state of patients who require an assistive device for walking. Given the variability that may exist among different individuals, this methodology tackles the issue from an individualized approach generating personalized models for each individual using the OC-SVM technique. The proposed methodology was validated in nine healthy people who had different simulated functional states, obtaining an accuracy in the range of 70–97%. In addition, a one-year longitudinal study was also carried out with three post-stroke individuals to validate the methodology in real cases, obtaining an average accuracy of 78%. Full article

Lower-limb rehabilitation exoskeletons have the potential to improve gait recovery after stroke by providing intensive and repetitive training. However, conventional control strategies often rely on fixed control parameters and exhibit limited adaptability to patient-specific characteristics, sensor noise, and dynamic uncertainties. This paper proposes an adaptive control framework that combines deep reinforcement learning (RL) with model-based impedance control for personalised lower-limb exoskeleton assistance. Patient-specific biological parameters are incorporated into the simulation environment and reward formulation to improve adaptability and robustness. Three state-of-the-art deep RL algorithms, Deep Deterministic Policy Gradient (DDPG), Twin Delayed Deep Deterministic Policy Gradient (TD3), and Soft Actor-Critic (SAC), are evaluated in a continuous control environment under varying signal-to-noise ratio (SNR) conditions ranging from 5 dB to noise-free conditions. Results demonstrate that TD3 achieves the most stable learning performance, obtaining a mean reward of −354.24 under noise-free conditions, while DDPG provides the highest joint-angle tracking accuracy with an RMSE of 0.0369 rad. SAC exhibits superior robustness in noisy environments, achieving the highest learning ratio of 0.51 at 5 dB SNR. Furthermore, the proposed personalised framework reduces tracking errors by up to 27% compared with non-personalised baseline approaches. The findings indicate that integrating patient-specific information with RL-based adaptive control can significantly enhance robustness, tracking performance, and personalisation in exoskeleton-assisted gait rehabilitation, providing a promising direction for future intelligent rehabilitation systems. Full article

Background/Objectives: Inclusion body myositis (IBM) causes progressive, quadriceps-predominant weakness that impairs mobility and increases fall risk, yet outcomes most relevant to independence remain unsynthesized for IBM. This review compiles direct IBM evidence across five domains (A1–A5): natural history, motor performance, falls and balance, sensory/peripheral-nerve function, and interventions. Methods: Eight databases and two trial registers were searched without date or language limits. Eligible studies enrolled adults with IBM based on recognized criteria reporting lower-limb strength or function, gait, transitional tasks, balance, falls, or intervention outcomes; mixed-myopathy cohorts required extractable IBM-specific data. Two reviewers independently screened, extracted, and appraised risk of bias, following PRISMA 2020. Results: Sixty-four studies were included; per-domain totals (A1: 16, A2: 9, A3: 6, A4: 2, and A5: 38) exceed 64 because studies may span domains. Quadriceps strength was the most sensitive progression marker, detected earlier by quantitative testing. Falls were near-universal and insufficiently managed. No drug showed convincing functional benefit in controlled trials, whereas exercise and orthotic/robotic assistance appeared to be safe in small studies. Sensory and peripheral-nerve dysfunction were common, but proprioceptive acuity and postural balance were unmeasured. Conclusions: IBM progression is best measured by quantitative quadriceps strength and function. Intervention evidence derives largely from small, uncontrolled and neutral trials. Primary myopathy is likely the principal driver of decline, but its downstream consequences—for balance, proprioception, and falls—remain underexplored and are the priority for future study. Full article

This paper proposes a hierarchical control architecture that is established at both lower limb joint and platform levels based on a simplified admittance model, where compliance is regulated through virtual mass and damping. At the lower limb joint level, admittance control governs lower limb motion tracking, while at the platform level it adjusts the omnidirectional mobile platform velocity in response to human interaction forces. Within this framework, a Sarsa-based reinforcement learning agent dynamically optimizes the parameters of a Sigmoid function using dual state inputs. Based on hip joint angle error and human–robot interaction force, the controller dynamically adjusts virtual mass and damping to optimize the trade-off between tracking error and dynamic compliance. The simulation and experimental results on the prototype system demonstrate that, compared with traditional Sigmoid parameter-tuned admittance control, the proposed approach significantly enhances gait smoothness (dimensionless squared jerk reduced by 65.62% and 36.74% for hip and knee joints), and increases human–robot interaction compliance (RMS interaction force was reduced from 3.2502 N to 2.5109 N; EPUD decreased from 12.14 to 9.53). Moreover, the proposed strategy achieves smaller maximum overshoot (0.45° vs. 0.9°) and faster settling time (2.6 s vs. 4.59 s). These findings indicate that integrating reinforcement learning with Sigmoid parameter adaptation provides a systematic and effective solution for adaptive compliance regulation in mobile exoskeleton systems, enhancing adaptability, safety, and functional relevance for stroke patients undergoing lower limb rehabilitation training. Full article

Accurate assessment of joint kinematics is fundamental to biomechanics, rehabilitation, and sports science. Although optical motion capture (OMC) remains the laboratory reference standard for biomechanical validation, its cost, infrastructure requirements, and limited applicability outside controlled environments restrict its broader use. Wearable inertial measurement units (IMUs) and vision-based markerless systems have consequently emerged as complementary alternatives, offering portability, reduced subject preparation, and applicability in ecological settings. Their rapid development, however, has not always been accompanied by an equally rigorous metrological interpretation of performance. This narrative review provides a comparative analysis of IMU- and vision-based approaches for joint kinematics estimation, focusing on biomechanical validation metrics and measurement error. Because the primary literature reports fundamentally different quantities under widely differing experimental conditions, evidence is presented stratified by outcome class, joint, plane of motion, task, and acquisition dimensionality, and values belonging to different outcome classes are not pooled. For sagittal-plane lower-limb angles during level walking in healthy adults, with careful sensor-to-segment calibration and an optoelectronic reference, IMU-based systems show the most consistent performance, with RMSE commonly between 3° and 6°. Vision-based systems achieve comparable accuracy for selected outcomes, particularly spatiotemporal gait parameters and sagittal-plane angles in controlled views, while degrading with occlusion, motion blur, and depth ambiguity. Accuracy is therefore not an intrinsic property of the sensing modality but of the entire measurement chain, including calibration, biomechanical modeling, acquisition geometry, and reporting conventions. Rather than ranking technologies by accuracy alone, the measurement requirements should be derived from the intended application. Full article

Background/Objectives: Sensor-based movement assessment is increasingly used to quantify movement execution quality and support feedback-guided rehabilitation exercise, particularly in home-based and remote rehabilitation contexts. However, exercise adherence, movement execution quality, and rehabilitation progress remain difficult to monitor continuously and objectively outside direct therapist supervision. This scoping review aimed to map the current applications of sensor-based biofeedback and movement-quality assessment systems for rehabilitation exercise and to identify evidence gaps. Methods: This review followed established scoping review methodology and PRISMA-ScR guidance. PubMed/MEDLINE, Web of Science Core Collection, and IEEE Xplore were searched, and Google Scholar was used for supplementary searching. English-language studies published from January 2014 to May 2026 were eligible if they involved rehabilitation-related populations, sensor-based movement assessment, biofeedback, or training guidance. Data were charted and narratively synthesized according to rehabilitation context, sensor technology, movement-quality metrics, computational approaches, feedback strategies, real-time or remote functions, and reported outcomes. Results: Fifty-five studies published between 2015 and 2026 were included. The evidence covered neurological, musculoskeletal and orthopedic, balance and vestibular, fall-prevention, home-based, and telerehabilitation applications. Technologies included inertial sensors, smartphones, vision/depth cameras, surface electromyography, pressure/force sensors, and multisensor systems. Movement-quality metrics included range of motion, postural stability, gait characteristics, loading, muscle activation, movement correctness, repetition count, and task completion quality. Feedback was visual, auditory, vibrotactile, app-based, avatar-based, therapist-facing, or remote-platform-based. Most evidence came from feasibility, technical validation, algorithmic validation, and small-sample clinical studies. Conclusions: Sensor-based systems may help translate rehabilitation exercise performance into quantifiable and feedback-enabled information. Future research should strengthen real-world validation, standardize task-specific movement-quality metrics, and clarify how feedback mechanisms can support individualized rehabilitation progression. Full article

Background: Quantitative assessment of dynamic balance remains challenging in clinical practice. Wearable inertial measurement unit (IMU)-based technologies offer an objective and accessible approach for monitoring postural control. This study aimed to develop and preliminarily validate a wearable IMU-based trunk sway monitoring system (SwayTracker) and investigate trunk sway characteristics in stroke patients walking with and without arm sling support. Methods: A sternum-mounted IMU system was developed to quantify dynamic trunk sway during walking. Fifteen healthy adults and fourteen stroke patients participated. The healthy participants established normative reference values, whereas the stroke patients completed walking trials with and without arm sling support. Trunk sway was quantified using anteroposterior (AP) and mediolateral (ML) deviations and a polar-coordinate-based sway model. Results: The healthy reference cohort exhibited a mean trunk sway magnitude (radius) of 8.20 ± 3.16°. The stroke patients demonstrated greater sway during unsupported (15.36 ± 5.83°) and sling-assisted walking (15.35 ± 4.59°). Although overall sway magnitude remained unchanged, the mean sway direction shifted from 69.49° to 110.20°, indicating a redistribution of trunk sway from the anterior-right toward the anterior-left quadrant. Forward sway remained the dominant AP component, whereas ML sway shifted from predominantly rightward to leftward with sling use. Conclusions: SwayTracker provides a feasible method for objective assessment of dynamic trunk sway during walking. The stroke patients exhibited increased sway magnitude and altered directional organization compared with healthy individuals. Arm sling use primarily modified ML postural compensation patterns rather than reducing overall trunk sway, highlighting the potential of wearable trunk sway monitoring for gait and balance assessment in neurological rehabilitation. Full article