Background Spinal cord injury affects approximately 9 million people worldwide, with incomplete injuries representing the majority of cases and offering greater potential for functional recovery. Overground robotic exoskeletons have emerged as promising rehabilitation tools, yet evidence regarding their effectiveness specifically during early inpatient rehabilitation remains inadequately synthesised. Objectives To systematically evaluate the effectiveness of overground robotic exoskeletons compared to conventional rehabilitation for improving walking function in adults with incomplete spinal cord injury during early inpatient rehabilitation. Methods This protocol follows PRISMA-P guidelines. Searches will be conducted across MEDLINE, Embase, CINAHL, Cochrane Library, and PEDro from inception through December 2025. Randomised and non-randomised controlled trials will be included. Two reviewers will independently screen studies using Covidence, extract data, and assess risk of bias using RoB 2 and ROBINS-I tools. Meta-analysis will be conducted where appropriate. Discussion This protocol establishes a rigorous framework for synthesising evidence on early exoskeleton-assisted rehabilitation, addressing a critical gap in clinical practice guidelines for spinal cord injury rehabilitation.

EnsNet: An ensemble environment classification in the application context of robotic leg prostheses and exoskeletons

In a rehabilitation exoskeleton, stable and safe operation is of central importance in rehabilitation training. This article develops a soft prescribed performance (SPP)-based reinforcement learning (RL) control method to address the conflict between performance constraints and system degradation, ensuring high accuracy and safe operation. First, a tunnel-type prescribed performance function is used to achieve faster convergence and smaller overshoot. Safety boundaries are used to define the tolerable error range, and an intermediate system links the safety and soft boundaries. The soft boundaries are dynamically adjusted to ensure safe operation by temporarily relaxing constraints during performance degradation. An RL approach based on an actor-critic (AC) structure is employed to handle unknown lumped disturbance. Theoretical analysis confirms the stability of the closed-loop system. Furthermore, a series of experiments is conducted on a self-built upper-limb rehabilitation exoskeleton robot driven by pneumatic artificial muscles to validate the effectiveness and robustness of the proposed method.

PurposeThis study aimed to investigate the safety, feasibility, and effect on hemiparetic muscle activity during sit-to-stand exercises using the lumbar type HAL in patients with acute-phase lower limb hemiparesis caused by ischemic stroke, cerebral hemorrhage, or postoperative brain tumor.MethodsThis randomized crossover study included twelve participants (ischemic stroke: 6, cerebral hemorrhage: 2, brain tumor: 4) for part 1, which assessed safety and comfort of sit-to-stand exercise using HAL, and ten participants (ischemic stroke: 4, cerebral hemorrhage: 3, brain tumor: 3) for part 2, which investigated the effect of HAL on muscle activities during sit-to-stand exercises. Participants performed either five sets (part 1) or a single set (part 2) of ten repetitions of sit-to-stand exercises under two conditions: with and without HAL assistance. Outcome included vital signs (blood pressure, heart rate, and percutaneous oxygen saturation), fatigue level in part 1. Muscle activity of bilateral gluteus maximus, biceps femoris, and vastus lateralis during the exercises were assessed in part 2. Outcomes were compared between the two conditions.ResultsThe results demonstrated that there were no significant changes in vital signs and fatigue between two conditions. Furthermore, HAL effectively increased muscle activity (μV) in the hemiparetic side vastus lateralis during the exercises (median, quartile range: 101.1, 55.8-125.9 vs. 79.9, 52.0-103.3, p = 0.03).ConclusionsThese findings suggest that the lumbar type HAL can be safely and feasibly used in patients with acute brain injury-related lower limb paresis, with potential effects on hemiparetic muscle activation. However, these results should be interpreted cautiously as exploratory observations.Registry informationClinical utility of cybernic system for patients with neurological diseases causing who need comprehensive nursing care (Clinical utility of cybernic system for patients with neurological diseases causing who need comprehensive nursing care (CUCSPND)), https://jrct.niph.go.jp/en-latest-detail/jRCTs052180074, jRCTs052180074.The effect of motor therapy using robot suit HAL for the patients with damaged brain (motor therapy using robot suit HAL), https://jrct.niph.go.jp/en-latestdetail/jRCTs052180223, jRCTs052180223.

Musculoskeletal disorders, particularly in the neck and back, are prevalent across various professions, stemming from prolonged static postures and awkward neck flexions. This study investigated the efficacy of a passive exoskeleton, designed to alleviate musculoskeletal neck and back strain, in a simulated neck flexion task. Ten participants performed tasks involving neck flexion at angles of 15°, 30°, 45°, and 60°, both with and without the exoskeleton. Additionally, the impact of using a headlight was evaluated at a 45° neck flexion angle. Wearable electromyography sensors were used to quantify muscle activity as an indicator of neuromuscular loading, while subjective discomfort was assessed using the Rate of Perceived Exertion scale, and endurance times were recorded. The results demonstrated significant reductions in neck and lower back muscle activity (median values up to 31.0%) and perceived discomfort (median values up to 50.0%), with the most improvements at 30° and 45° neck flexion angles. Participants reported 50% higher endurance time when using the exoskeleton. Minimal benefits were observed at 15° flexion, likely due to reduced musculoskeletal demand at this angle. These findings highlight the potential of exoskeletons as an ergonomic intervention to mitigate neck and back strain in occupations where high degrees of neck flexion are prevalent.

The rapid advancement of technology is transforming the landscape of sports medicine, offering new opportunities to enhance injury prevention, diagnosis, treatment, and rehabilitation. This article explores the integration of cutting-edge technologies in sports recovery processes, motivated by the growing demand for faster, more effective, and personalized approaches to athlete care. It reviews key innovations including wearable devices, artificial intelligence (AI), virtual and augmented reality (VR/AR), robotics, regenerative medicine, and telemedicine. These technologies are reshaping traditional practices by enabling real-time data collection, predictive injury modeling, remote monitoring, and biologically driven treatment protocols. The article highlights how AI-powered analytics can personalize rehabilitation plans, while robotic systems and exoskeletons contribute to improved motor recovery. VR/AR technologies offer immersive therapy experiences, enhancing neuromuscular engagement and patient motivation. Furthermore, developments in regenerative medicine, such as stem cell therapies and 3D bioprinting, demonstrate promising results in tissue repair and recovery acceleration. While these innovations offer significant benefits, challenges such as cost, accessibility, data privacy, and ethical considerations must be addressed. Overall, this review underscores the transformative potential of new technologies in sports medicine and emphasizes the need for continued interdisciplinary research to optimize their implementation and ensure equitable access. The results suggest a paradigm shift in recovery practices, paving the way for more responsive, data-driven, and patient-centered approaches to sports healthcare.

Post-stroke hemiplegia of the upper extremities continues to pose a significant therapeutic hurdle. Contralesional uncrossed corticospinal pathways (CST) are involved in the recovery processes. We test the safety, and preliminary efficacy of targeted upregulation of uncrossed CST excitability through self-modulation of cortical activities via noninvasive brain-machine interaction training (Registered with the University Hospital Medical Information Network: UMIN000017525). In this single-arm prospective trial, eight individuals with persistent severe post-stroke motor disability voluntarily actuated their affected shoulder using a brain-computer interface (BCI) bridging the contralesional motor cortex (M1) and an exoskeleton robot. While patients attempted to elevate the affected arm, scalp electroencephalogram (EEG) signals over the contralesional M1 were processed online to provide them with feedback on M1 excitability. Here we show that the BCI reconstructs neural pathways, allowing arm elevation without any adverse effects. As evidenced by an increase in primary outcome measure (Fugl- Meyer Assessment, p < 0.05, d = 1.24), seven days of consecutive system use results in rapid, sustained, and clinically significant improvement in motor function when removed from the system and promotes contralesional M1 functional remodeling. This closed-loop system is safe, feasible, and a promising intervention that recruits intact neural resources to allow patients to recover upper-extremity motor abilities.

Posterior communicating artery (PСоA) aneurysms represent a serious medical problem associated with a high risk of rupture leading to severe neurologic deficits, disability, and, in some cases, death. Despite the advances in microsurgical and endovascular treatment, the recovery of patients requires long-term, comprehensive, and multidisciplinary neurorehabilitation aimed at correcting motor, cognitive, and psychosocial disorders. The present review examines current approaches to rehabilitation, including early initiation, as well as the use of specialized scales and neuropsychological tests in combination with promising methods of hyperbaric oxygenation and neuromodulation. Particular attention is paid to artificial intelligence (AI) technologies in neurorehabilitation: the use of adaptive game systems, robotic exoskeletons, brain-computer interfaces, and gamification to personalize and increase the effectiveness of recovery programs. The integration of AI into the rehabilitation process opens up new opportunities for improving the functional outcomes and quality of life in patients with PСоA aneurysms. However, further research and a systematic approach in care management are required.

Background. Total endoprosthesis, including total knee arthroplasty (ТКА), is the main treatment method for severe forms of osteoarthritis. However, the quality of life in patients after surgery largely depends on the management of the rehabilitation process. The development of new rehabilitation methods and a need to study the efficiency of traditional ones enhance the relevance of the systematization and analysis of new studies on ТКА rehabilitation. Objective: to identify key trends in rehabilitation of ТКА patients and to consider current rehabilitation methods with proven efficiency. Material and methods. Literature search was conducted in scientific databases and electronic libraries: PubMed/MEDLINE, Google Schoolar, and eLibrary. A total of 2860 publications were identified, including 70 fully eligible and considered in the review. Sources were searched and selected based on PRISMA guidelines. Results. The current trends in ТКА rehabilitation involve early initiation of recovery procedures, digitally managed late postoperative period, and a personalized approach. The use of proven methods ensures rehabilitation according to the specified trends. These include exercise, biofeedback, and physical (cryotherapy, neuromuscular electrical stimulation, and electroacupuncture) therapy, as well as the latest technologies: robotic exoskeletons and virtual reality devices. Specially designed mobile applications and remote monitoring tools enable efficient telerehabilitation. This personalized approach contributes to patients’ satisfaction with the recovery process. Conclusion. The development of modern rehabilitation methods ensures comfortable recovery procedures for ТКА patients and their return to full life within a relatively short time.

Non-primary motor areas, including dorsal premotor cortex (PMd), ventral premotor cortex (PMv), and posterior parietal cortex (PPC), contribute to movement planning, but how these regions differentially shape kinematic features of goal-directed movements, and how this specialization is associated with functional connectivity within the frontoparietal network, remains of interest, particularly in relation to recovery after stroke. We used functional magnetic resonance imaging (fMRI), transcranial magnetic stimulation (TMS), and kinematic assessments to explore how these areas influence reaching performance in neurologically intact adults. Participants performed a goal-directed planar reaching task using the KINARM exoskeleton robot. Brief TMS pulse trains were initiated before movement onset to perturb cortical activity at subthreshold and suprathreshold intensities targeting bilateral PMd, PMv, and dorsomedial superior parietal lobule (SPL) within PPC. Resting-state fMRI quantified functional connectivity among these regions to assess whether connectivity explains stimulation-induced kinematic changes. Relative to the control target within the postcentral sulcus (PCS), subthreshold stimulation of contralateral PMd and PMv reduced reach efficiency and smoothness, while suprathreshold stimulation of contralateral PPC increased deviation error and reduced smoothness. Among ipsilateral targets, PMd showed consistent TMS-induced effects, and was the only target where resting-state connectivity predicted behavioral response. Stronger interhemispheric connectivity in the primary motor cortex and weaker interhemispheric PPC connectivity were associated with greater reductions in straightness and smoothness during subthreshold ipsilateral PMd stimulation. We found that perturbation of premotor and parietal targets led to distinct kinematic effects that varied by site, intensity, and laterality, with premotor stimulation showing the most consistent disruptions at subthreshold intensity and bilateral effects, whereas parietal effects were observed primarily for contralateral stimulation at suprathreshold intensity, and differences in network organization explain variability in behavioral response. Identifying contributions of cortical areas and connectivity patterns may help personalize interventions after stroke. Trial Registration: This study was registered at ClinicalTrials.gov under ID NCT04286516.

Sliding mode-based actuator fault reconstruction and fault-tolerant control of lower limb rehabilitation exoskeleton robots

The triceps surae muscle serves as the primary power source for ankle push-off during normal walking and contributes to improved gait economy. However, insufficient power output from the muscles around the ankle joint leads to increased metabolic cost. In this study, a pneumatically driven biarticular knee-ankle exoskeleton robot was designed based on the biomechanical principles of ankle push-off, employing artificial pneumatic muscles to mimic the synergistic actuation characteristics of the soleus and gastrocnemius muscles. Using the indirect calorimetry device K5, assistance performance experiments of the exoskeleton were conducted under different walking speeds (0.2, 0.4, and 0.6 m/s), slopes (level ground, 5° uphill, and 5° downhill), and assistance parameters (low, medium, and high magnitude). The results showed that compared with walking unpowered exoskeleton, medium-intensity (2.25 bar) assistance reduced the metabolic cost by approximately 22% (2.23 ± 0.47 W kg −1 ) during walking at 0.2 m/s on level ground. During walking at 0.6 m/s on a 5° uphill slope, high-intensity (3.0 bar) assistance reduced the metabolic cost by approximately 15% (4.08 ± 1.04 W kg −1 ). The effectiveness of assistance varied significantly across terrains under the same assistance intensity. This study provides valuable insights into the exoskeleton adaptability used on diverse terrains and the bionic design of exoskeletons.

Abstract Chitin is a valuable bioresource, a sustainable biomaterial that is obtained from marine shell wastes. Chitin has been utilized in various diverse applications, cutting edges across human, societal, and environmental benefits. From chitin, numerous other derivatives have been derived, extrapolating the utility of this sustainable material. Voluminous research publications exist concerning chitosan and its derivatives, but very few are available that exclusively deal with chitin and its applications. Chitin nanocomposites helped overcome the inherent limitations of chitin, predominantly its bioavailability. The present review specifically reviews the contributions of chitin nanocomposites/metal nanocomposites for biomedical applications. With chitosan and the other derivatives being projected predominantly for drug delivery applications, this review presents a streamlined targeted survey on the milestones achieved by chitin nanocomposites and metal nanocomposites. A comparison between chitin and chitosan metal nanocomposites in drug delivery applications has also been presented and inferences discussed. The dearth in availability of prominent research studies for chitin nanocomposites as well as their corresponding metal nanocomposites, has been highlighted. The fact that the potential of chitin, nanoforms, and nano‐metal composites are being undermined, because of inadequate research evidence, which has led to preliminary conclusions, has also been discussed.

In recent years, rehabilitation exoskeletons have become a rapidly developing technology for rehabilitating and aiding people with various types of hand functional disabilities. However, one of the intractable problems is the lack of methods that can effectively convert the intentions of wearers to the motions of exoskeletons. In order to systematically understand the working principles and operation methods of some advanced control strategies, and to explore their potential. This review introduces three existing and modern control schemes in soft robotic hand exoskeleton from the perspective of their target, analysis methods and operating ways, with practical applications in every strategy. Then, the remaining context explores individual advantages and drawbacks to discuss the opportunities and challenges met in some state-of-the-art patterns that relate the bio-signal to the movement of robotic hand exoskeletons. Finally, the conclusion deduces that the prospective control schemes are the development of hybrid control architectures that synergistically integrate complementary advantages from multiple methodologies. Such hybrid approaches are expected to enhance the rehabilitation efficacy of robotic prosthetic devices significantly, optimizing patient-specific functional recovery outcomes and providing flexible rehabilitation plan.

Abstract This study prepared wholly deacetylated, low-molecular-weight chitooligosaccharides (COS) through cellulase disintegration of chitosan sourced from shrimp and crab to produce (SCO) and CCO, respectively. FTIR analysis showed that the employed conditions (55 °C, pH 5.2 in 24 h) enabled nearly complete deacetylation, 98.8% and 100%, with average molecular weights of 1.288 and 0.467 kDa, respectively. The ESI/MS findings revealed that the resulting COS consisted of monomers and polymers of D-glucosamine (1–6 units). The COS demonstrated a progressive increase in water solubility, culminating at 88% and displayed exceptional bioactivities, particularly in their scavenging activity against DPPH at concentrations of 1–6 mg/mL. SCO produced scavenging rates equivalent to 54.72 ± 2%-60.18 ± 1%, against 59.11 ± 1%- 65.29 ± 1% in the case of CCO. The antibacterial efficacy COS (1200 µg/mL) revealed maximum inhibition zones of 26 ± 1 and 22 ± 1 mm against Bacillus cereus and Escherichia coli. The inclusion of COS in yogurt and orange juice reduced the total aerobic count by about 0.7 ± 0.1 log CFU/mL and garnered excellent sensorial acceptability ratings, exceeding 4 ± 0.0. This study is innovatively using cellulase, a cost-effective, widely available enzyme, to produce fully deacetylated, low-molecular-weight COS from marine shell waste. These ameliorated COSs offer a sustainable alternative to synthetic preservatives, particularly in yogurt and orange juice. They present a potential in green food preservation and an eco-friendly approach to transforming seafood waste into high-value bioactive compounds.

To investigate the clinical efficacy of exoskeleton-assisted training on lower limb functional recovery in the early rehabilitation phase following anterior cruciate ligament reconstruction (ACLR). Between February 2024 and August 2024, 20 patients who were 2 to 3 weeks after anterior cruciate ligament reconstruction (ACLR) were selected, including 13 males and 7 females, with ages ranging from 18 to 45 years old (28.5±6.2) years old. These patients were divided into a conventional rehabilitation group and an exoskeleton-assisted group, with 10 patients in each group. The conventional rehabilitation group consisted of 6 males and 4 females, with a mean age of (24.60±1.78) years old, and received routine rehabilitation training. The exoskeleton-assisted group included 7 males and 3 females, with a mean age of (25.20±1.93) years old;on the basis of conventional rehabilitation, exoskeleton robot assistance was applied during gait training for this group. Both groups underwent a 4-week intervention program, five days per week. Lower limb function was assessed pre-intervention and at 4 weeks post-intervention using the Lysholm knee score, visual analogue scale (VAS) for pain, knee flexion range of motion (ROM), and center of pressure (COP) displacement area. Postoperative wound healing was satisfactory with no complications such as infection or re-injury. All 20 patients completed the 4-week intervention and assessments. After 4 weeks of intervention, the Lysholm scores of both groups were significantly higher than those before intervention;the score of the exoskeleton-assisted group (82.30±4.15) was higher than that of the conventional rehabilitation group (72.60±4.88), with a statistically significant difference(P<0.001). The VAS score of the exoskeleton-assisted group (2.38±0.52) was lower than that of the conventional rehabilitation group(3.45±0.69), with a statistically significant difference (P=0.001);the knee flexion ROM of the exoskeleton-assisted group(124.50±5.34)° was greater than that of the conventional rehabilitation group (115.80±5.76)°, with a statistically significant difference(P=0.002);the COP displacement area of the exoskeleton-assisted group (6.28±0.94) cm2 was smaller than that of the conventional rehabilitation group (8.45±1.12) cm2, with a statistically significant difference (P<0.001). Exoskeleton-assisted robotic training more effectively improves early knee function, reduces pain, increases joint range of motion, and enhances balance control in patients following ACLR. It may be applied as an optimized approach for early postoperative rehabilitation.

Against the backdrop of intensifying global population aging, lower-limb exoskeleton robots serve as core devices for rehabilitation and power assistance. Their control accuracy and motion smoothness rely on precise dynamic models. However, parameter uncertainties caused by variations in human lower limbs, assembly errors, and wear pose a critical bottleneck for accurate modeling. Aiming to achieve high-precision dynamic modeling for a two-degree-of-freedom lower-limb exoskeleton, this paper proposes a parameter identification method named Tent-GA-GWO. A dynamic model incorporating joint friction and link inertia was constructed and linearized. An excitation trajectory based on Fourier series, conforming to human physiological constraints, was designed. To enhance algorithm performance, Tent chaotic mapping was employed to optimize population initialization, a nonlinear control parameter was used to balance search behavior, and genetic algorithm operators were integrated to increase population diversity. Simulation results show that, compared to the traditional GWO algorithm, Tent-GA-GWO improved convergence efficiency by 32.1% and reduced the fitness value by 0.26%, demonstrating superior identification accuracy over algorithms such as GA and LIL-GWO. Validation on a physical prototype indicated a close agreement between the computed torque based on the identified parameters and the actual output torque, confirming the method’s effectiveness and engineering feasibility. This work provides support for precise control of exoskeletons.

With the gradual development of exoskeleton robot hardware, the research focus should be transferred to the control model. The current field of exoskeleton control systems is undergoing a paradigm shift from traditional human-computer interaction to human-computer symbiosis. This research focuses on the control model and control strategy of an exoskeleton robot. Through a systematic review of the literature, this paper divides the current control model into four clear control paradigms: brain-less control, perceptual control, predictive control, and symbiotic control, and prospects the future human-computer interaction mode of exoskeleton. At the same time, the core characteristics, typical cases, and limitations of each paradigm are analyzed, and the four trends of human-computer symbiosis are summarized. The future development path of exoskeleton based on artificial intelligence, brain-computer interface, and new sensing technology is outlined. Brainless control can not adapt to individual differences and real-time changes, ignoring the real-time state of people. Perceptual control cannot predict or look forward to motor intention. Predictive control is complex in calculation, highly dependent on the model and data, and mostly in the laboratory stage. In the future, a close coupling and bidirectional communication man-machine relationship will be formed to form a man-machine intelligent unity.

To meet the requirements of the lower limb exoskeleton robot working in coordination with the human body and improve the human-machine interaction performance, a lower limb motion intention recognition method based on the dual-stage joint optimization of the Gated Recurrent Unit (GRU) neural network by the Grey Wolf Optimization (GWO) and the Adaptive Boosting (AdaBoost) algorithm is proposed, and the GWO-AdaBoost-GRU intention recognition model is constructed. The Surface electromyography signals of six lower limb movements are collected and processed respectively by the CEEMDAN-WT joint denoising, activity segment extraction, and feature extraction, and the feature vector dataset is constructed as the model input. To comprehensively verify the performance of the GWO-AdaBoost-GRU model, it is compared with the GRU and GWO-GRU models, and an application verification is carried out by building a lower limb exoskeleton rehabilitation system. The experiments show that the average recognition accuracy of the GWO-AdaBoost-GRU model is 95.5%, which is 8.1% higher than that of the GRU model and 3.2% higher than that of the GWO-GRU model. Moreover, in the practical application of the lower limb rehabilitation institution, the GWO-AdaBoost-GRU intention recognition model has high accuracy, can accurately recognize the movement intentions of the subjects, and complete the designated rehabilitation movements in conjunction with the rehabilitation system, demonstrating excellent application performance.

In the field of robotic exoskeleton control, it is critical to accurately predict the intention of the user. While surface electromyography (EMG) holds the potential for such precision, current limitations arise from the absence of robust EMG-to-torque model calibration procedures and a universally accepted model. This paper introduces a practical framework for calibrating and evaluating upper-limb EMG-to-torque models, accompanied by a novel nonlinear model. The framework includes an in situ procedure that involves generating calibration trajectories and subsequently evaluating them using standardized criteria. A comprehensive assessment on a dataset with 17 participants, encompassing single-joint and multi-joint conditions, suggests that the novel model outperforms the others in terms of accuracy while conserving computational efficiency. This contribution introduces an efficient model and establishes a versatile framework for EMG-to-torque model calibration and evaluation, complemented by a dataset made available. This further lays the groundwork for future advancements in EMG-based exoskeleton control and human intent detection.