Search Results (1,193)

Stroke is a leading cause of chronic disability, with heterogeneous sensorimotor impairments that are not well captured by standard clinical assessments. While advanced motion capture and robotic systems provide precise measurements, they are not scalable for widespread clinical use. We developed C-MORE (Computer Vision for Movement Observation and Recovery Enhancement), a smartphone-based framework that uses computer vision and machine learning to automatically score the Box and Blocks Test (BBT) and extract quantitative kinematic metrics. The system combines hand tracking with a custom machine learning (ML) architecture to identify valid block transfers and segment task phases. We evaluated C-MORE in 7 individuals with chronic stroke and a cohort of 10 healthy adults. The system achieved 99.0% agreement with ground-truth scoring, with errors below clinically meaningful thresholds. Kinematic measures derived from the system were sensitive to stroke-related impairments, including reduced movement velocity and increased task duration in affected limbs. Exploratory analyses indicated that grasp-related metrics, particularly the ratio of grasp-to-transfer duration, were correlated with independent measures of proprioception. These findings demonstrate that smartphone-based computer vision can provide accurate, scalable assessment of upper-extremity function. C-MORE offers a practical approach for enhancing clinical evaluation and enabling more precise, individualized rehabilitation strategies. Full article

Objective: To systematically evaluate the rehabilitation effect of brain-computer interface (BCI)-controlled hand robot training on post-stroke motor functions, especially upper limb functions. Methods: PubMed, Embase, Web of Science, Cochrane Library, CNKI, SinoMed, WanFang Data, and VIP Database were searched from inception to 13 March 2026. Randomized controlled trials (RCTs) with dose-matched designs were included, where the test group underwent BCI-controlled hand robot training and the control group received either pure hand robot training or routine rehabilitation. Meta-analysis was performed on RevMan 5.4. Results: Totally 11 RCTs involving 380 patients were included. Compared with hand robot training alone, BCI-controlled hand robot training significantly improved Fugl-Meyer Assessment for Upper Extremity (FMA-UE) scores (MD = 4.87, 95% CI: 1.04 to 8.69) and FMA-UE proximal scores (MD = 4.44, 95% CI: 0.15 to 8.74), and significantly reduced finger flexor spasticity (MD = −0.44, 95% CI: −0.68 to −0.21), but showed no significant difference in distal upper limb motor function or Action Research Arm Test (ARAT) scores. Compared with routine rehabilitation, BCI-controlled hand robot training significantly improved FMA-UE scores (MD = 6.55, 95% CI: 3.49 to 9.61). Conclusions: BCI-controlled hand robot training can effectively improve overall upper limb and proximal motor function after stroke and alleviate finger flexor spasticity, but the evidence for distal hand function and long-term efficacy remains limited. Full article

Robust exercise recognition is essential for robot-assisted lower-limb rehabilitation, where misclassifications of sensor-derived movements can degrade therapy execution and supervision. This study proposes a novel hybrid weighted stacking ensemble to increase the efficiency of the intelligent module of the LegUp parallel robotic system for lower limb rehabilitation. The approach combines a Residual Multilayer Perceptron (ResMLP) and an optimized Kernel Extreme Learning Machine (KELM), where model hyperparameters are tuned using Optuna and the base-model probability outputs are fused through optimized weighting and a meta-learner. Experiments were conducted on a five-class dataset built from nine IMU orientation features acquired from three sensors placed on the healthy limb. Four meta-learners were evaluated (Logistic Regression, Random Forest, Gradient Boosting, and AdaBoost), with AdaBoost providing the best overall performance. To further assess the robustness and generalization capability of the proposed approach, a 5-fold cross-validation procedure was performed for the ResMLP, KELM, and the hybrid ensemble models. The proposed stacking hybrid ensemble consistently surpassed the performance of the strongest individual classifiers as well as the original LegUp Multilayer Perceptron model. These results indicate that combining residual learning with kernel-based classification in a weighted stacking framework yields a stable and high-performing solution for multi-class rehabilitation exercise recognition. Full article

The Compliant Ankle Rehabilitation Robot (CARR) is actuated by four Pneumatic Artificial Muscles (PAMs). These actuators mimic biological muscles, making them highly applicable in robotic systems designed to apply guiding motion to a human joint, but have complex nonlinear dynamic properties making accurate tracking control difficult. This paper presents intelligent modelling and control methods to improve the CARR’s function in a rehabilitation setting. Using the Particle Swarm Optimisation (PSO) algorithm, dynamic models of the actuators are calculated. Two model setups are proposed, a single model and dual model. Model Predictive Control (MPC) was then implemented using these models and experimentally compared with Proportional Integral Derivative (PID) and Iterative Learning Control (ILC). The results showed that PID control was less accurate than the developed control schemes, with evidence of significant chattering, as well as both over- and undershooting the setpoint. ILC performed accurately, but the required learning period and some evidence of overfitting impacted the overall performance. Single-model MPC had low error values on the X axis of the CARR and maintained the most consistent 0 displacement when an axis was intended to stay motionless. Dual-model MPC had the lowest error values on the Y axis, the smoothest motion with the least chattering, and the best performance when both axes were in motion, but showed evidence of overshooting. Based on these results, both MPC implementations have proven to be successful and suitable for future work, and the PSO modelling method is able to produce suitably accurate models for the application. Full article

Background/Objectives: Recovery of upper limb function after stroke is highly heterogeneous, and accurate prediction of clinically meaningful functional transition remains a major challenge in rehabilitation medicine. We developed and temporally validated machine learning (ML)-based prognostic models for predicting transition from non-functional movement to functionally usable upper limb capacity in patients undergoing intensive inpatient rehabilitation during the early subacute phase of stroke. Methods: This retrospective cohort study included 960 patients with ischemic or hemorrhagic stroke admitted to a tertiary rehabilitation center between 2010 and 2025. Three functional recovery outcomes were defined: motor impairment recovery, defined as Fugl-Meyer Assessment for Upper Extremity score ≥ 32; gross manual dexterity recovery, defined as Box and Block Test score ≥ 2 blocks/min; and functional pinch strength recovery, defined as pinch strength ≥ 1.1 kgf. Multidimensional predictors spanning demographic, clinical, neurophysiological, neuroimaging, and rehabilitation-related domains were integrated. Four ML algorithms were evaluated using stratified 5-fold cross-validation and temporal validation in a chronologically independent cohort (2024–2025). Models were developed under two tracks: Track A, incorporating only baseline variables available at admission (primary prognostic model), and Track B, additionally incorporating cumulative rehabilitation-related variables (exploratory). Results: Random Forest demonstrated the best overall performance. During temporal validation, models achieved AUROC of 0.800 for motor impairment recovery, 0.958 for gross manual dexterity recovery, and 0.888 for functional strength recovery. Baseline motor severity and corticospinal tract integrity were the dominant biological determinants of recovery. Earlier rehabilitation initiation and greater upper-limb robot-assisted therapy exposure were also associated with improved outcomes; however, these findings should be interpreted as observational associations subject to treatment-selection bias rather than evidence of causal effects. Conclusions: Probabilistic ML prediction integrating neural reserve and rehabilitation-related exposure variables can support individualized precision rehabilitation planning and improve functional outcome stratification in early subacute stroke. Full article

The aging population and the high incidence of neurological disorders have driven an increasing demand for lower-limb motor dysfunction rehabilitation. Traditional rehabilitation methods suffer from limitations such as low efficiency and a lack of personalization. Lower-limb rehabilitation exoskeleton robots have emerged as a critical solution, with human–robot intelligent fusion serving as the core theoretical framework and technological pathway for performance enhancement. From the unique perspective of human–robot intelligent fusion, this paper systematically reviews the application and recent advances of artificial intelligence in three key aspects—intention perception, intelligent control, and human–robot integration—based on a layered architecture of “fusion perception, fusion decision-making, and fusion execution”. The definition, connotations, and realization mechanisms of human–robot intelligent fusion are clarified. Furthermore, this review analyzes the fusion mechanisms, applicable scenarios, and technical characteristics of different AI technologies and summarizes the human–robot intelligent fusion modes and clinical application status of representative products such as EksoNR, MyoSuit, and AiLegs. In addition, key challenges are identified from the perspectives of fusion generalization capabilities, the trade-off between real-time performance and robustness, algorithm interpretability, and multimodal deep fusion mechanisms. This paper provides a systematic theoretical reference and technical roadmap for establishing a unified human–robot intelligent fusion framework for lower-limb rehabilitation exoskeletons. Full article

The growing integration of rehabilitation robotics and artificial intelligence has created new opportunities for developing control strategies that better support clinicians during patient therapy. This study investigates machine learning and deep learning approaches for classifying upper limb motion using encoder-based biomechanical data, with the goal of identifying a model suitable for implementation in a rehabilitation exoskeleton. Several classical algorithms such as k-Nearest Neighbors, Random Forest, multiclass logistic regression, XGBoost, and an SVM classifier were evaluated alongside three deep learning architectures: convolutional layers, GRU and LSTM units. Models were trained and tested on two types of datasets using both standard cross-validation and leave-one-subject-out validation. The analysis included assessments of class separability, signal features’ importance, and comparative performance based on F1-score, accuracy, and confusion matrices. Results showed notable differences between validation strategies, with LOSO evaluation revealing limitations of the available dataset and emphasising the need for broader data collection. Overall, the findings indicate that, in the LOSO evaluation of the five-class multi-subject dataset—the most clinically realistic validation scenario—the LSTM-based model achieved the highest generalisation performance (accuracy 92.8%, macro-F1 0.927), supporting its suitability for integration into exoskeleton control systems aimed at detecting and mitigating compensatory movements. Full article

Biomechanics sits at the interface of engineering and medical sciences, offering essential insight into how tissues, organs, and biological systems respond to mechanical loading. This review brings together recent advances in musculoskeletal and cardiovascular biomechanics, illustrating how experimental techniques, computational modeling, and multiscale analysis are used to characterize load transfer, tissue deformation, fatigue, and injury mechanisms. In musculoskeletal applications, predictive simulations, wearable sensing technologies, and neuromechanical assessment tools support improved injury prevention, rehabilitation planning, and assistive device development. In the cardiovascular domain, patient-specific modeling, fluid–structure interaction analyses, and advanced imaging approaches clarify how hemodynamics, vessel wall mechanics, and device–tissue interactions influence disease progression, implant performance, and therapeutic outcomes. Emerging technologies including artificial intelligence, machine learning, digital twin frameworks, biofabrication, soft robotics, and self-powered sensing are enabling data-driven, real-time, and personalized interventions that connect mechanistic understanding with clinical practice. Despite these advances, challenges remain in accounting for individual variability, integrating multiscale data, and translating computational predictions into clinically validated solutions. By emphasizing interdisciplinary strategies that unite biomechanics, computational analytics, and innovative device engineering, this review outlines a pathway toward predictive, patient-centered healthcare and next-generation therapeutic and rehabilitation solutions. Full article

Robot-assisted telerehabilitation (RAT) combines rehabilitation robotics with digital health workflows to extend access to upper-limb (UL) therapy after stroke. Mixed reality (MR) may support therapist–patient interaction and task visualization; however, early-stage systems require rigorous evaluation of safety and usability before deployment in the home. In a formative, mixed-methods usability study conducted in a controlled setting using a telerehabilitation workflow, six individuals post-stroke (≥3 months) and six occupational therapists (OTs) completed a single supervised session with a desktop-mounted end-effector type therapeutic robot (iTbot) integrated with Microsoft HoloLens 2. Participants performed structured passive and active UL exercises while therapists supervised and interacted with the system via the MR control interfaces. Safety was evaluated by documenting observed adverse events and safety-stop activations. Usability and user experience were assessed using the System Usability Scale (SUS), study-specific satisfaction questionnaires (reported with scale ranges), and semi-structured follow-up interviews analyzed using thematic analysis. All participants completed the session without observed adverse events or safety-stop activations. Overall usability was favorable, with a mean (SD) SUS total score of 78.3 (15.9) out of 100 (stroke: 74.2 [18.1]; occupational therapists: 82.5 [13.5]). Qualitative feedback indicated that MR was perceived as engaging and intuitive by many users, while also identifying implementation needs relevant to real-world telerehabilitation, including clearer onboarding, simplification of certain MR interactions, and improved physical interfaces (e.g., handle options). Therapists highlighted workflow considerations for remote supervision and patient independence. Together, these findings support progression to multi-session, in-home studies to quantify remote assistance needs, technical reliability, adherence, and clinical outcomes. Full article

In recent years, hand exoskeleton robots have attracted extensive attention from researchers and practitioners due to their potential to rehabilitate, assist, and enhance hand movements, particularly for stroke patients. With an ageing population increasingly affected by strokes, there is a growing demand for patient-centred interventions which place less demand on clinicians, especially wearable devices that can enhance hand function. Advances in artificial intelligence have opened new avenues for developing more reliable and adaptive assistive systems. This study presents a systematic literature review, following the PRISMA protocol on the design elements of hand exoskeleton robots, acknowledging the emerging perspectives on AI integration and ethical considerations. The study provides a comprehensive foundation for future research and development in rehabilitation technologies by systematically synthesising the current mechanical architecture, actuation, sensors, material, weight, and cost aspects of soft hand exoskeleton robots for rehabilitation. The results show important patterns and trade-offs in various design dimensions, providing useful information to direct the development of more accessible and efficient rehabilitation solutions in the future. Full article

The number of patients with lower-limb dysfunction is increasing each year due to aging, illness, accidents, and other factors. Without timely rehabilitation and rapid recovery of walking function, further physical and mental deterioration may be accelerated, potentially leading to long-term bedriddenness. This study discusses gait training in rehabilitation therapy from the perspectives of kinesiology, cognitive science, walking function, and safety, and an omnidirectional gait training robot was developed. This study proposed a control system construction method for an omnidirectional gait training robot based on both prescription-based training and autonomous training. In the prescription-based training system, the target values are derived from the training prescription, and the control objective is to guide the patient to walk along the robot’s prescribed path and speed. In the autonomous training system, the target values are automatically generated based on the patient’s walking intentions, and the control objective is for the robot to safely follow the patient’s movement. A necessary condition for robot-assisted autonomous gait training is effective fall prevention. A fall prediction strategy based on foot position information and handrail pressure data was developed. Using this strategy, the robot can predict falls immediately before they occur, similar to a physical therapist, thereby reducing the risk of falls during gait training. Experimental results demonstrate the feasibility of implementing this strategy. Full article

Artificial intelligence (AI) is reshaping medical rehabilitation by advancing from isolated assistive technologies toward data-driven program management. Beyond established applications in robotics and virtual reality, AI enables multimodal data integration, predictive analytics, adaptive therapy optimization, and real-time monitoring across rehabilitation domains. This review synthesizes 61 peer-reviewed studies to examine how AI supports the management, planning, and evaluation of rehabilitation programs. The evidence indicates strong technical maturity at the device and session levels, particularly in robotic control and wearable monitoring, whereas longitudinal program orchestration and system-level coordination remain at an emerging stage. Machine learning, reinforcement learning, computer vision, and time-series models facilitate patient phenotyping, therapy personalization, and prognostic modeling. However, their scalability is constrained by limited interoperability, heterogeneous outcome measures, and insufficient multicenter validation. A structured six-layer management architecture is proposed to conceptualize AI as an integrated orchestration framework. Advancing toward scalable and trustworthy rehabilitation ecosystems will require interoperable infrastructures, longitudinal validation, and embedded ethical and explainability mechanisms. Full article

Neurological disorders often cause severe upper limb motor impairments that restrict independence and quality of life. Robot-assisted rehabilitation enables high-intensity, task-oriented, and quantifiable training. One key feature, gravity compensation (GC), reduces the muscular effort needed to lift the limb and supports voluntary movement by offsetting the weight of the arm. This systematic review aimed to identify the types of GC strategies used in upper limb rehabilitation robots and assess clinical evidence on their effectiveness for improving motor outcomes. A search of PubMed, Scopus, Web of Science, and IEEE Xplore (January 2005–May 2025) identified 60 eligible studies: 23 describing GC implementation and 40 reporting clinical results. GC was implemented into exoskeletons, end-effectors, and sling-suspension systems through passive mechanical designs or active, model-based, and adaptive control algorithms. However, few studies reported key technical parameters such as controller algorithms, loop frequency, or tuning procedures, and only one addressed the control system stability. Clinically, GC-assisted training improved arm movement and range of motion, with greater effects in participants with higher impairment levels. However, the functional gains were modest and not superior to conventional or other robotic therapies. Substantial heterogeneity in training protocols and participants’ demographics further limits direct comparison among GC strategies. Overall, the relative effectiveness of robot-assisted GC across devices remains unclear. Standardized reporting and more clinical trials are needed to compare GC strategies within and between different types of robots. Full article

Background/Objectives: Robot-assisted therapy (RAT) can deliver repetitive, feedback-enriched upper-limb practice after stroke, but evidence comparing RAT with dose-matched conventional occupational therapy (COT) under routine inpatient conditions—and concurrent neurocognitive data—remains limited. We compared motor recovery between an end-effector RAT-based program (30 min RAT plus 30 min COT) and dose-matched COT alone in subacute stroke survivors, with neurocognitive outcomes prespecified as exploratory endpoints. Methods: In this single-center retrospective non-randomized cohort study, adults with first-ever ischemic or hemorrhagic stroke who completed routine baseline and week−4 assessments received 4 weeks of upper-limb rehabilitation: combined RAT plus COT (60 min daily) or COT alone (60 min daily). The primary outcome was the week-4 Fugl–Meyer Assessment–Upper Extremity (FMA-UE) motor score adjusted for baseline. Primary inference used covariate-adjusted linear regression on outcome-specific complete cases, with a prespecified stabilized inverse probability of treatment weighting (IPTW) average treatment effect analysis as the sensitivity test. Secondary and exploratory endpoints were interpreted descriptively; Benjamini–Hochberg false discovery rate (FDR) control and multiple imputation were applied as supportive analyses. Results: The analytic cohort comprised 65 patients (RAT, n = 33; COT alone, n = 32). Both groups improved over 4 weeks, but the RAT group had worse baseline upper-limb motor status. The adjusted between-group difference for the week-4 FMA-UE motor score was non-significant (adjusted mean difference, 4.39; 95% confidence interval (CI), −2.43 to 11.21; p = 0.203), and the stabilized IPTW estimate was concordant (β = 2.17; 95% CI, −3.63 to 7.98; p = 0.464). In unadjusted analyses, the FMA-UE motor gain was larger after RAT than COT alone (14.70 ± 15.53 vs. 7.91 ± 9.42), and only the RAT group exceeded the prespecified 12.4-point clinically important threshold; this signal attenuated after adjustment. No secondary or exploratory endpoint remained significant after FDR control. Multiple imputation for the primary endpoint was concordant with the complete-case result (pooled β = 4.52; 95% CI, −1.91 to 10.94; p = 0.168). Conclusions: End-effector RAT did not demonstrate adjusted superiority over dose-matched COT alone for upper-limb motor recovery. The larger unadjusted FMA-UE gain should be interpreted as a descriptive impairment-level signal rather than as evidence of comparative efficacy. Neurocognitive results were exploratory; the retrospective non-randomized design, baseline imbalance, differential missingness, and unavailable confounder data require cautious interpretation. Full article

Spinal cord injury (SCI) is an acute, devastating neurologic condition that results in permanent progressive motor deficits, sensory disturbances, and autonomic dysfunctions, which limit function, participation, and quality of life. Although substantial progress has been made during the last several decades for both early trauma care and rehabilitation protocols following SCI, long-term neurological recovery remains unpredictable and often incomplete. This manuscript summarizes mechanistic and clinical evidence regarding robotic-assisted rehabilitation (RAR) and spinal neuromodulation (SN), which have been published since 2010 until the present time in a structured narrative review of the literature on these two emerging areas for neurorehabilitation after SCI. RAR provides high-intensity, task-specific training that consistently results in improvements in functional outcomes such as balance, coordination, and independence; however, its impact is limited when it comes to walking speed or voluntary motor control. SN (particularly epidural stimulation) can activate the residual neural pathways to standing up and stepping even after a complete injury but effects are typically stimulus dependent, with heterogeneous clinical results that often lack strong long-term evidence due in part to variability in patient selection, stimulation parameters and rehabilitation protocols. However, there is emerging mechanistic data supporting combining modulation of excitability through SN approaches along with structured sensorimotor training as an approach for enhancing recovery. Collectively, these findings support a shift toward more physiology-driven neurorehabilitation strategies and the need for future research to improve clinical translation and outcome predictability by patient stratification using standardized intervention protocols that include longitudinal evaluation. Full article