搜索结果：Journal of neuroengineering and rehabilitation

Stroke is one of the leading causes of mortality and long-term disability worldwide. In recent years, integrated rehabilitation models that combine virtual reality (VR) technology with standardized exercise therapy have emerged, demonstrating promising potential in improving recovery outcomes. This bibliometric review systematically analyzes global literature on virtual reality combined with exercise therapy for stroke rehabilitation to map the knowledge landscape, identify research hotspots and evolutionary trends, and inform future research, clinical practice, and policy. Relevant studies on VR combined with exercise therapy for stroke rehabilitation were retrieved from the Web of Science database, covering the period from database inception to 2025. Bibliometric and visualization analyses were conducted using CiteSpace and VOSviewer to assess publication trends, country, institutional contributions, authors and co-cited authors networks, highly cited references, core journals, and the evolution of research hotspots. A total of 1,687 articles were identified, showing a steady upward publication trend. China ranked first in publication volume, while the United States had the highest total citation count. Researchers such as Calabrò, De Luca, and Naro from the IRCCS Centro Neurolesi in Messina, Italy, made notable contributions, particularly in VR-robotics combined rehabilitation. The Journal of NeuroEngineering and Rehabilitation published the largest number of articles in this field. Keyword burst analysis indicated two distinct phases: before 2021, research primarily focused on conventional rehabilitation methods and clinical trials; after 2021, attention shifted towards the integration of emerging technologies in stroke rehabilitation, including machine learning and immersive VR, reflecting growing scholarly interest in novel rehabilitation strategies. This study provides a comprehensive bibliometric analysis of VR combined with exercise therapy in stroke rehabilitation, identifying key research hotspots, emerging trends, and existing limitations. The findings could offer theoretical insights and data-driven evidence to support future research and clinical applications in this field.

Robot-assisted therapy for poststroke rehabilitation of the upper limb is rapidly spreading. The need comes from the reduced number of therapists and the aim of defining a more involving therapy for the patients. Previous studies report the effectiveness of robotic therapy to provide intensive, repetitive and task-specific rehabilitation, as well as the ability to provide different modalities of training. However, it is not always clear how these modes are implemented and how they are defined, since different labels are sometimes used. As a consequence, it is difficult to define a universal protocol to follow. This leads to non-comparable outcomes, making even harder for the therapists to understand which is the most efficient way of administering therapy. The proposed work aims at putting together available information reported in literature, linking two main variables influencing rehabilitation, i.e., poststroke stage and training modality, with the purpose of updating the state of the art, categorising and analysing the modalities involved, extracting the most effective relationships and approaches in terms of results reported in the scientific community. Scopus was chosen as reference database for the systematic review. The studies refer to the last decade, that is from the year of the latest review published in relation to the topic of interest. Studies clearly referencing training modalities and poststroke stages were included. The assistive modality is the one that catches more attention in the scientific community, highlighting the tendency to prefer approaches in which the patients are more actively involved. In terms of relevance inferable from clinical scales reported in the included studies, the assistive modality appears to be the most effective in the chronic phase, active-assistive approaches during the subacute one, whereas no significant conclusions can be drawn for the acute stage. For what concerns robotic devices, some considerations can be drawn as well: exoskeletons are applied during the chronic phase predominantly, whereas end-effectors during the subacute one. No significant distinctions are detected in the acute stage. Improvements in ADLs are mostly achieved in experiments involving exoskeletons, but studies show that subjects may also benefit from end-effectors, when applied in earlier recovery stages. It is evident that some connections are present between training modalities and recovery stages, influencing the outcomes of experimental trials. Evaluation metrics exploited in tests report enhanced outcomes when the association between training modality and poststroke stage is optimized. Nevertheless, future developments will possibly extend this study to other factors that may have influenced outcomes, such as intensity of the exercises, frequency and duration of the therapy, and impairment severity. Moreover, a deeper analysis, incorporating investigations on daily clinical practice, would help to identify of the most effective approaches.

Due to their high power-to-weight ratio, modular and reconfigurable architectures, and inherent compliance, cable-driven rehabilitation robots (CDRRs) provide safe, lightweight, backdrivable solutions for gait and movement rehabilitation. However, they continue to face unique control challenges due to cable properties and user variability. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, this systematic review explores control strategies for lower-limb CDRRs from the past decade. Out of the 968 studies initially identified, 70 met the selection criteria and were classified into six categories: position and velocity, force- and torque-based, compliance-based, model-based and optimal, learning-based and intention-informed, and hierarchical frameworks. Our analysis revealed a chronological evolution from traditional classical control toward more personalized, adaptive, learning-based, and intention-driven methods. Impedance and admittance control remain fundamental for ensuring safety, while newer approaches enable user-specific and environment-responsive assistance. This review proposes a unified hierarchical framework linking high-level intent detection to low-level actuation providing researchers and developers with a structured understanding of the control landscape for cable-driven lower-limb exoskeletons in healthcare and beyond. Control strategies were also linked to clinical outcomes to relate them to functional improvements across patient populations. Advancing CDRRs will require unified, multi-layer architectures that couple constraint-aware model-based control with adaptive and intention-driven learning to achieve safe, scalable, and clinically meaningful rehabilitation.

Commercially-available microprocessor-controlled prosthetic knees are unable to fully replicate the biomechanical function of the missing biological limb. While powered prostheses have the capacity to restore joint level kinetics, current systems rely on intrinsic control schemes that do not allow the user to volitionally modulate movement under neural commands. This limitation may compromise functional performance and hinder prosthetic embodiment, the sense that the device is part of the user’s body. In a case study on one test participant, we evaluate the functional and perceptual benefits of a bone-anchored, neurally-controlled knee prosthesis by comparing it to the participant’s microprocessor-controlled prosthesis. We conducted a within-subject study on an individual with a transfemoral amputation, with an osseointegrated implant and surgically reconstructed agonist–antagonist muscle pairs. We tested a neurally-controlled powered knee and conventional microprocessor knee across a set of activities, including seated volitional control tasks, sit-to-stand transitions, squatting, level-ground walking, stair ascent, and uninstructed standing. Performance metrics included knee kinematics, prosthesis-generated mechanical power, and functional outcomes such as gait speed, stair ascent time, and weight-bearing symmetry derived from ground reaction forces. Functional mobility and control were complemented by self-reported embodiment, assessed through a questionnaire targeting agency, ownership, and body representation. The neurally-controlled prosthesis enabled intuitive and responsive control. Compared to the subject’s prescribed prosthesis, the prosthesis yielded improved temporal gait symmetry during walking (symmetry index: 0.93 vs. 0.59, with 1 indicating perfect stance time symmetry), increased prosthetic-side weight-bearing during sit-to-stand and squatting, and successful execution of a step-over-step stair ascent strategy—an outcome not achievable with the subject’s prescribed device. Embodiment scores were consistently higher with the neurally-controlled prosthesis compared to the prescribed device across multiple domains, including agency, ownership and body representation. This study is the first to directly compare a prescribed microprocessor knee with a bone-anchored, neurally-controlled powered prosthesis. By combining osseointegration, surgically reconstructed agonist–antagonist muscle pairs, and powered actuation, the system improved gait symmetry, greater prosthetic-side loading, and step-over-step stair ascent. These results demonstrate the novelty and promise of integrating surgical and mechatronic innovations to restore both functional mobility and embodied control after transfemoral amputation. This study was approved by the Institutional Review Board at MIT (Protocol No. 2503001589). The online version contains supplementary material available at 10.1186/s12984-026-01921-y.

The Cybathlon is an international competition in which individuals with physical disabilities complete activities of daily living using advanced assistive technologies, including robotic devices. International championships were held in Zurich in 2016, 2020, and 2024, with up to eight disciplines that included races using powered prosthetic limbs, exoskeletons, and wheelchairs. Between 2015 and 2024, more than 120 teams from over 30 countries participated in Cybathlon competitions and affiliated events. The aim of this work is to systematically assess the Cybathlon’s broader impact over its first decade, with particular attention to public visibility, scientific dissemination, and the translation of assistive technologies toward academic, clinical, and industrial application. To this end, we combine a longitudinal analysis of media coverage (2014–2025), a bibliometric review of scientific publications, and a survey of participating teams. A total of 6,944 media items were identified, predominantly from online sources (73.6%), with the highest coverage originating from Switzerland (26%), followed by the USA, Italy, and Germany. By January 2025, 297 scientific publications referred to the Cybathlon, with author affiliations most frequently from Switzerland, Italy, the USA, and China. Among participating teams, 79% reported accelerated prototype development due to Cybathlon participation, and over one-third attributed new educational materials, clinical methods, or collaborations to their involvement. These effects were frequently linked to early and sustained user integration, iterative testing under realistic conditions, and close interaction between engineers, clinicians, and end users. Participation directly contributed to the founding of three companies and supported five additional ventures. Additionally, at least 21 established companies participated during the various editions. Overall, the Cybathlon emerges as a unique, competition-driven innovation ecosystem that not only increase public awareness of assistive technologies but also shapes research priorities, support user-centered design practices, and facilitates pathways toward clinical adaptation and commercialization. The online version contains supplementary material available at 10.1186/s12984-026-01988-7.

Children with Unilateral Cerebral Palsy (UCP) are characterized by significant upper limbs (ULs) impairment that hinders daily activities and participation, requiring accurate assessment. Although motor impairment mainly affects one side, significant evidence suggests that the less affected side (LAS) may also present impairments. Considering the growing interest in pediatrics technologies, a complementary approach combining standardized clinical scales and robotic devices may support a more objective and quantitative evaluation of ULs performance with high precision and objectivity. This study aimed to investigate the feasibility of a planar end-effector robotic device in children and adolescents with UCP and to explore whether robotic ULs indices show associations with clinical measures. Twenty-eight children and adolescents with UCP (mean ages: 10.90 ± 3.32 years; 8 males, 20 females) underwent a single-session protocol comprising: (i) clinical assessment based on ULs classifications - Manual Ability Classification System (MACS), House Functional Classification System (HFCS) - and clinical standardized scales - Melbourne Assessment 2 (MA2), Box and Block Test (BBT) -; (ii) robotic assessment protocol with MOTORE with more and less affected side (MAS and LAS respectively), and (iii) feasibility questionnaires completed by both children and clinicians to evaluate usability and acceptability. Feasibility data were analysed as raw scores (mean value, SD, range) and percentages. Two different MANOVA analyses were applied to robotic parameters, including limb condition (LAS vs. MAS), hand dominance, and age as predictors, with HFCS or MACS levels analysed in separate models to avoid conceptual overlap and collinearity. Finally, Spearman’s r correlation coefficient assessed correlations between clinical and robotic measures. Feasibility questionnaires showed positive results, supporting the system’s user-feasible design. Across the MOTORE tasks, MANOVA analyses showed that HFCS or MACS levels and limb severity condition (MAS vs. LAS) were the main factors affecting robotic performance, with age influencing selected parameters, while hand dominance had no significant effect. Finally, correlation analyses further revealed moderate to strong associations between MOTORE’s robotic parameters and clinical scores (r range: 0.375–0.643). This paper highlights the feasibility of MOTORE robotic system in pediatric assessment setting both for users and clinicians. The observed associations suggest that the system may provide complementary quantitative information related to ULs functions. As these findings are exploratory, further studies are required to establish validity, reliability, and responsiveness before clinical integration of the system to support objective and quantitative data to guide individualized rehabilitation strategies for children with UCP. Trial registration: ClinicalTrial.gov: NCT06012617 and NCT06666829. The online version contains supplementary material available at 10.1186/s12984-026-01950-7.

Patients implanted with the PRIMA photovoltaic subretinal prosthesis in geographic atrophy report form vision with the average acuity matching the 100 μm pixel size. Although this remarkable outcome enables them to read and write, they report difficulty with perceiving faces. Despite the pixelated stimulation, patients report seeing smooth patterns rather than dots. This paper provides a novel, non-pixelated algorithm for simulating prosthetic vision the way it is experienced by PRIMA patients, compares the algorithm's predictions to clinical perceptual outcomes, and offers computer vision and machine learning (ML) methods to improve face representation. Our simulation algorithm (ProViSim) integrates a grayscale filter, spatial resolution filter, and contrast filter. This accounts for the limited sampling density of the retinal implant (pixel pitch), as well as the reduced contrast sensitivity of prosthetic vision. Patterns of Landolt C and faces created using this simulator are compared to reports from actual PRIMA users. To recover the facial features lost in prosthetic vision due to limited resolution or contrast, we apply an ML facial landmarking model, as well as contrast-adjusting tone curves to the image prior to its projection onto the photovoltaic retinal implant. Prosthetic vision simulated using the above algorithm matches the maximum letter acuity observed in clinical studies, as well as the patients' subjective descriptions of perceived facial features. Applying the inversed contrast filter to the image prior to its projection onto the implant and accentuating the facial features using an ML facial landmarking model helps preserve the contrast in prosthetic vision, improves emotion recognition and reduces the response time. Spatial and contrast constraints of prosthetic vision limit resolvable features and degrade natural images. ML based methods and contrast adjustments prior to image projection onto the implant mitigate some limitations and improve face representation. Even though higher spatial resolution can be expected with implants having smaller pixels, contrast enhancement still remains essential for face recognition.

To investigate the effects of multisensory-integrated virtual reality (VR) training on gait adaptability and its regulatory mechanisms on the somatomotor network (SMN) in patients with stroke. In this randomized controlled trial, 68 patients with stroke were allocated to a VR group (multisensory-integrated VR training) or a control group (conventional rehabilitation). Both groups received 30-minute sessions, 5 days/week for 4 weeks. The primary outcome was gait adaptability assessed by the Dynamic Gait Index (DGI). Secondary outcomes included the Timed Up and Go Test (TUGT), Berg Balance Scale (BBS), and Fugl-Meyer Assessment for Lower Extremity (FMA-LE). Functional near-infrared spectroscopy (fNIRS) measured resting-state functional connectivity within the SMN and task-evoked activation during stepping and obstacle crossing. The VR group showed significantly greater improvements than the control group in DGI total score (P&#x2009;=&#x2009;0.010), TUGT (P&#x2009;=&#x2009;0.005), and BBS (P < 0.001 ). fNIRS analysis revealed that the VR group exhibited significantly greater increases in task-evoked activation in the right posterior parietal cortex (PPC) and supplementary motor area (SMA) during stepping (P&#x2009;=&#x2009;0.029 and P&#x2009;=&#x2009;0.032, respectively), and in the right SMA during unaffected-limb obstacle crossing (P&#x2009;=&#x2009;0.048). Resting-state functional connectivity analysis showed significantly enhanced connections within the SMN, including left SMA-right PPC and right PPC-left dorsolateral prefrontal cortex (DLPFC) (both P&#x2009;<&#x2009;0.05). Correlation analyses revealed that increased right SMA activation during obstacle crossing was positively correlated with TUGT improvement (r&#x2009;=&#x2009;0.590, P&#x2009;=&#x2009;0.001), while enhanced right PPC-left DLPFC connectivity was positively correlated with DGI improvement (r&#x2009;=&#x2009;0.403, P&#x2009;=&#x2009;0.041). Multisensory-integrated VR training was associated with improvements in gait adaptability and balance in patients with stroke. The underlying mechanisms may involve enhanced activation in the SMA and PPC, along with changes in functional connectivity within the SMN and between the SMN and cognitive control networks. However, given that the significant improvements in DGI, TUGT, and BBS were not sustained under the most conservative assumptions about missing data, these findings should be considered preliminary and warrant confirmation in studies with lower attrition rates. Chinese Clinical Trial Registry, ChiCTR2500111919 (retrospectively registered). Registered 7 November 2025. Available from https//www.chictr.org.cn (registration number ChiCTR2500111919). Protocol The full trial protocol is available from the corresponding author upon reasonable request.

Complex walking tasks require enhanced cognitive control to meet sensorimotor integration demands, yet the neurophysiological mechanisms underlying how the post-stroke brain adaptively adjusts to such challenges remain insufficiently understood. The spectral distribution of electroencephalography (EEG) oscillations and the functional network construction method offer a window into real-time neural resource allocation and network reorganization during complex walking. One baseline control (steady-state level walking) and two challenging conditions were implemented: an asymmetrical board-walking task requiring dynamic balance control, and a visual-deprivation walking task that increased reliance on proprioceptive and vestibular feedback. Scalp EEG was recorded simultaneously during walking. Fifty-seven post-stroke participants completed all experimental conditions. A three-way repeated-measures ANOVA was applied to examine spectral power changes across tasks, frequency bands, and brain regions. The brain functional connectivity was computed using the weighted phase lag index. Additionally, a two-way repeated-measures ANOVA was conducted to analyze changes in brain network topological properties across different frequency bands and task conditions. Compared to steady-state walking, both the balance-challenged and visually deprived walking tasks consistently suppressed delta power in temporal regions and theta power across occipital and parietal areas (p < 0.05). Beta power was enhanced in temporal and parietal regions during both tasks (p < 0.05), while occipital alpha power increased specifically during visually deprived walking task (p < 0.05). Gamma-band activity remained unmodulated across conditions. The two challenging walking tasks increased functional connectivity in the alpha, beta, and gamma frequency bands but reduced theta-band connectivity. Graph-theoretical analysis demonstrated that both tasks elicited higher clustering coefficients in the alpha band (p < 0.05). In contrast, only visually deprived walking task led to a significant reduction in the delta-band clustering coefficient (p < 0.05). In response to locomotor challenges, people with stroke showed a neural reorganization involving suppressed low&#x2011;frequency oscillations and enhanced mid&#x2011;to&#x2011;high&#x2011;frequency activity, which may reflect a re&#x2011;allocation of neural resources toward cognitive&#x2011;motor integration demand. Trial registration The study protocol was registered on ClinicalTrials.gov (No. NCT06395142).

Non-invasive brain stimulation (NIBS), including transcranial direct current stimulation (tDCS) and repetitive transcranial magnetic stimulation (rTMS), is a potential treatment for chronic low back pain (CLBP). However, its efficacy and safety remain inconclusive. In this systematic review and meta-analysis, the clinical benefits of NIBS for pain relief, functional improvement, and quality of life (QoL) in patients with CLBP were evaluated. Five databases (PubMed, Web of Science, Cochrane Library, Embase and Scopus) were searched to identify randomised controlled trials (RCTs) of NIBS for CLBP. The final review included 21 RCTs involving 744 participants. The included studies assessed pain intensity, disability, and QoL outcomes, along with adverse events. Risk of bias (RoB) was evaluated using the RoB-2 tool, and meta-analyses were conducted using the R platform. NIBS was found to significantly reduce pain intensity (standardised mean difference [SMD]&#x2009;=&#x2009;&#x2009;&#x2212;&#x2009;0.85 [&#x2212;&#x2009;1.21,&#x2009;&#x2212;&#x2009;0.49]) and disability (SMD&#x2009;=&#x2009;&#x2009;&#x2212;&#x2009;0.52 [&#x2212;&#x2009;0.89,&#x2009;&#x2212;&#x2009;0.16]) compared with sham or control interventions. Subgroup analyses revealed that tDCS yielded broader benefits for both pain (SMD&#x2009;=&#x2009;&#x2009;&#x2212;&#x2009;0.80) and disability (SMD&#x2009;=&#x2009;&#x2009;&#x2212;&#x2009;0.59), while rTMS had the largest effect on pain reduction (SMD&#x2009;=&#x2009;&#x2009;&#x2212;&#x2009;1.09). tDCS targeting the primary motor cortex (M1) showed the strongest effects on pain relief (SMD&#x2009;=&#x2009;&#x2009;&#x2212;&#x2009;0.95) and functional improvement (SMD&#x2009;=&#x2009;&#x2009;&#x2212;&#x2009;0.59), while dorsolateral prefrontal cortex stimulation also yielded significant benefits for both pain (SMD&#x2009;=&#x2009;&#x2009;&#x2212;&#x2009;0.64) and disability (SMD&#x2009;=&#x2009;&#x2009;&#x2212;&#x2009;0.65). There is very low to moderate evidence that NIBS can help relieve pain and reduce disability in individuals with CLBP. The online version contains supplementary material available at 10.1186/s12984-026-01886-y.

Foot drop in patients with stroke compromises gait and mobility. Active ankle movement with real-time ultrasound-imaging visual feedback (UVF) is found to be a clinically effective rehabilitation protocol; however, its underlying neural mechanisms in poststroke brains remains unclear. This functional near-infrared spectroscopy (fNIRS) study aimed to investigate UVF's neuromodulatory effects during ankle dorsiflexion (ADF) execution. Fifteen post-stroke patients (SP) and fifteen healthy controls (HC) performed active ADF tasks with and without UVF. The fNIRS measured the change of oxyhemoglobin (&#x394;HbO) and deoxygenated hemoglobin (&#x394;HbR) over the dorsolateral prefrontal cortex (DLPFC), the primary motor cortex (M1), and the premotor cortex/supplementary motor areas (PMC/SMA). Functional connectivity (FC) between these regions was calculated. The muscle morphology parameters [muscle thickness (MT), pennation angle (PA), fascicle length (FL)] were quantified via ultrasound imaging during movement execution. Two-way mixed ANOVA was used to assess the main and interaction effects of "task" and "group" on the local cortical activation levels and the FC among pairwise regions of interest (ROI). Correlations between muscle parameters and M1-related fNIRS data were analyzed. UVF did not increase &#x394;HbO at any ROIs; instead, it reduced &#x394;HbR in the I PMC/SMA in both groups. UVF also decreased the interhemispheric DLPFC connectivity in both groups, based on residual FC data. Regardless of whether UVF was provided or not, SP consistently showed stronger I PMC/SMA - I DLPFC connectivity than HC. SP also exhibited a UVF&#x2011;specific increase in the I PMC/SMA - I M1 connectivity relative to HC. During UVF and with comparable number of ADF, SP showed reduced PA and FL modulation than that of HC. Among MT, FL, and PA, only FL showed a trend-level positive association with interhemispheric M1 connectivity in SP, on top of the local M1 activation. Generally, UVF during ADF was associated with the reduced bilateral prefrontal coupling and the enhanced ipsilesional premotor-motor connectivity in post-stroke patients. Ankle muscle FL displayed certain level of task sensitivity, and showed a preliminary association with interhemispheric motor coordination, implying for possible network&#x2011;level influences during UVF&#x2011;guided motor execution.

Accurately distinguishing minimally conscious state plus (MCS+) from minimally conscious state minus (MCS-) is critical for prognosis and treatment planning. Microstate analysis decomposes multichannel electroencephalography (EEG) into a sequence of brief, relatively stable scalp electric-field topographies, offering a unique spatiotemporal perspective on brain activity. Yet applications of microstate methods to the assessment of disorders of consciousness remain scarce. Moreover, most state-of-the-art studies focus on characterizing the complexity of microstate sequences, while conventional complexity measures overlook transitions between microstates. To address this gap, we propose Microstate Permutation Lempel-Ziv Complexity (MS-PLZC), an extension of Lempel-Ziv complexity that explicitly encodes ordinal permutation information to more sensitively capture the temporal organization of microstate sequences. Resting-state EEG was recorded from 45 individuals with disorders of consciousness (15 unresponsive wakefulness syndrome, 15 MCS-, 15 MCS+) and 15 neurologically healthy controls. MS-PLZC, conventional microstate LZC, spectral power, sample entropy, and classical LZC were calculated and statistically compared. These features were assessed using a nested leave-one-out cross-validated (LOOCV) SVM with exhaustive hyper-parameter search. Both MS-LZC and MS-PLZC showed statistically significant group differences (Kruskal-Wallis test: MS-LZC: H&#x2009;=&#x2009;26.92, p&#x2009;<&#x2009;0.0000, &#x3b7;&#xb2;=0.2099; MS-PLZC: H&#x2009;=&#x2009;35.11, p&#x2009;<&#x2009;0.0000, &#x3b7;&#xb2;=0.2816), with MS-PLZC exhibiting greater statistical power. Notably, MS-PLZC successfully distinguished between MCS- and MCS+ patients (p _adj&#x2009;<&#x2009;0.05) with a large effect size (Cliff's Delta = -0.6178), whereas MS-LZC demonstrated only a medium effect size (Cliff's Delta = -0.3067). In the machine-learning analysis MS-PLZC achieved the highest leave-one-out accuracy (0.733) and ROC-AUC (0.733). These results indicate that MS-PLZC sensitively captures subtle shifts in microstate dynamics and offers a reliable single-feature discriminator of MCS+ versus MCS-, with translational potential for detecting key recovery windows during routine assessment of consciousness.

Stroke patients with hemiplegia often show inefficient gait patterns, including reduced knee flexion during the swing phase, which may increase fall risk. Post-stroke gait frequently involves merged muscle synergies that affect lower limb kinematics. However, it remains unclear how muscle synergy merging and fractionation relate to knee flexion during the swing phase. Therefore, this study aimed to examine the association between knee flexion during the swing phase and muscle synergy merging and fractionation patterns in patients with stroke. The study comprised 21 stroke patients with hemiplegia. Surface electromyography was recorded from eight lower-limb muscles on the paretic side during comfortable gait. Maximum knee flexion angle (MKFA) during the swing phase was measured using a markerless motion capture system. Using non-negative matrix factorization, the number of muscle synergies, their spatiotemporal structure were calculated. Participants were classified into a low-synergy group (LS; n&#x2009;=&#x2009;5; one or two synergies) or a high-synergy group (HS; n&#x2009;=&#x2009;16; three synergies). Group comparisons of MKFA during the swing phase were performed. Furthermore, we investigated whether muscle synergies of the HS group could be fractionations of those of the LS group. The HS group showed significantly greater MKFA compared with the LS group (p&#x2009;=&#x2009;0.032). In the HS group, the ankle plantar flexors constituted an independent muscle synergy, whereas in the LS group, these muscles had high weightings within a muscle synergy associated with load response. Furthermore, the independent muscle synergies observed in the HS group were shown to be fractionated from the merged muscle synergies present in the LS group. Our results showed that merged muscle synergies were associated with reduced MKFA during the swing phase, whereas an independent synergy involving the plantar flexors was associated with greater knee flexion. These findings suggest that fractionation of the plantar flexor synergy may be important for improving knee kinematics after stroke and could inform targeted rehabilitation strategies. Given the relatively small and imbalanced sample size, cautious interpretation of the findings is warranted. Further studies with larger, balanced samples are needed to further strengthen the evidence for these findings.

Successful participation in the CYBATHLON depends not only on technical innovation but also on effective, systematic training of pilots with spinal cord injury. Current research primarily focuses on clinical rehabilitation, while standardized, task-specific training protocols for applied competition settings are lacking. To address this gap, this methodology paper presents a scientifically grounded training concept for exoskeletons, focusing on the qualification of trainers as key facilitators who adapt exercises and mediate between human and machine. A modular 12-week training program with 38 units was developed and implemented through a Moodle-based e-learning platform, demonstrating the practical applicability of the proposed methodological framework. The program progresses from basic movement exercises to competition-specific obstacle tasks from the CYBATHLON Exoskeleton Race. The curriculum integrates theoretical content on safety and exoskeleton handling with practical training scenarios. Trainer qualification forms the central element, enabling individualized adaptation of exercises to pilot capabilities and exoskeleton requirements. The concept allows for scalable application and provides a consistent framework that can be updated according to future CYBATHLON regulations and technological developments. The developed user-centered and digitized training program provides an effective structure for preparing trainers and pilots for the CYBATHLON. Combining modular design, digital learning tools, and individualizable adaptation, it establishes a transferable framework applicable to other exoskeleton systems. Beyond its practical use, the framework introduces a replicable methodology for developing and implementing training concepts in human&#x2013;machine interaction, forming a foundation for evidence-based and transferable training standards in assistive robotics.

Lower-limb exoskeletons are a useful tool in rehabilitation settings as they can provide customized assistance to individuals during functional exercises. These approaches typically rely on state-machine-based control with impedance controllers tailored to different locomotion phases, ensuring appropriate assistance across various activities and environments. However, these methods necessitate lengthy calibration procedures, as many impedance parameters need to be fine-tuned to provide appropriate assistance for various activities (e.g., overground walking, ramps, and stairs). This study presents three contributions: (1) a state-machine-based control strategy for partial assistance lower-limb exoskeletons, (2) a computational method to extract reference trajectories from a benchmark dataset (Camargo et al. in J Biomech 119:110320, 2021), enabling the identification of state-machine controller parameters and simplifying calibration procedures and (3) a dataset of 19 healthy individuals walking in five walking conditions (overground walking, upstairs, downstairs, up ramps, and down ramps) using either the state-machine approach or a transparent controller. The state-machine controller produced in average more negative interaction power ([Formula: see text] W/kg) compared to transparent control ([Formula: see text] W/kg), indicating greater user assistance. Preferred walking speed was notably faster with the state-machine controller, particularly on level ground, ramps and stairs ascent (25&#x2013;32% increase). Kinematic analysis revealed closer alignment to able-bodied gait patterns with the state-machine controller, suggesting improved gait quality. At the same time, the dataset of the collected locomotion activities (dataset link) will constitute a new benchmark dataset for locomotion. In this work, we presented and evaluated a novel state-machine-based control strategy for partial-assistance lower-limb exoskeletons. In this approach, reference trajectories are extracted from a benchmark dataset, simplifying calibration procedures. Additionally, we provide a dataset of 19 healthy individuals using two exoskeleton controllers. The proposed controller will be applied to patient populations, while the dataset will serve as a valuable resource for advancing robust and effective control mechanisms through machine learning techniques. The online version contains supplementary material available at 10.1186/s12984-026-01917-8.

Postural collapse and increased reliance on accessory respiratory muscles during meals can compromise ventilatory efficiency, particularly in individuals with respiratory impairment. Forward elbow-supported sitting is commonly used to unload the upper limbs; however, this posture often induces trunk flexion and cervical extension, which may adversely affect respiratory mechanics. To date, the effects of lateral armrest support on posture and respiratory function during seated eating have not been quantitatively investigated. To examine the effects of lateral armrest support on thoracic alignment, cervical posture (cervical inclination angle), shoulder muscle stiffness (upper trapezius and middle deltoid), vital capacity, and subjective comfort during seated eating in healthy young adults, compared with unsupported sitting and anterior elbow-supported sitting. Forty healthy young adults completed three randomized sitting conditions: (A) unsupported sitting, (B) anterior elbow-supported sitting, and (C) lateral armrest-supported sitting using a side-mounted armrest. Thoracic kyphosis index, cervical inclination angle, muscle stiffness of the upper trapezius and middle deltoid, vital capacity (VC), and subjective comfort were assessed. Data were analyzed using Friedman tests, with post-hoc pairwise comparisons performed using the Durbin-Conover test with Bonferroni correction. Thoracic kyphosis index differed significantly across conditions (p = .002), with both forward elbow-supported and lateral armrest-supported sitting showing lower values than unsupported sitting. Cervical inclination angle also differed significantly (p < .001), with forward elbow-supported sitting demonstrating greater cervical extension than the other conditions. Upper trapezius muscle stiffness was significantly reduced in the lateral armrest-supported condition compared with unsupported and forward elbow-supported sitting (both p < .01). Vital capacity was significantly greater in the lateral armrest-supported condition than in unsupported sitting (p = .003). Subjective comfort ratings were highest in the lateral armrest-supported condition (p < .001). Lateral armrest support demonstrated biomechanical advantages in healthy young adults, including improved spinal alignment, reduced shoulder girdle loading, and greater vital capacity compared with unsupported sitting. These findings provide preliminary mechanistic insight and may inform future investigations in clinical populations; however, direct extrapolation to individuals with respiratory or neurological impairment requires further study.

User preference is important for improving the acceptance of assistive robotic devices. However, it is not clear what factors influence user preferences or how it relates with physiological variables, such as muscle activity. Three participants with transfemoral amputation walked on a treadmill at a self-selected speed using a knee-ankle prosthesis. In four trials conducted on the same day, participants identified their preferred push-off parameters by self-tuning the magnitude and timing of the assistance using a 2D grid interface through a self-exploration method. They were blinded to the control parameters and relied on their perception of the assistance to guide the tuning. During these trials, muscle activity was recorded from four muscles in the sound limb. Muscle intensity and muscle synergies were calculated to explore the relation between muscle activity and user preference. While the participants generally preferred low to medium magnitudes of assistance, their preferred settings were not consistent across trials. In most trials these preferred settings were associated with lower muscle intensity in most muscles and better motor coordination. These findings suggest that physiological metrics, such as muscle intensity and muscle synergies, may influence the selection of preferred prosthesis settings. Even though user preference is likely influenced by multiple factors, such as gait symmetry, understanding the role of muscle activity and its relationship with preference could support the development of more personalized assistive robotic devices. However, further research is needed to confirm these observations and better explain how muscle activity contributes to participants&#x2019; preferred settings. The online version contains supplementary material available at 10.1186/s12984-026-01931-w.

Chronic neuropathic pain is a complex experience that poses a major challenge in personalized treatment. Identifying objective biomarkers of pain modulation is critical to validate emerging non-pharmacological therapies with reliable endpoints, overcoming the limitations of simplified subjective scales. Here, we introduce a multimodal monitoring framework that integrates behavioral, sensory, and cortical assessments to provide a comprehensive evaluation of a multisensory neurostimulation treatment combining immersive VR with targeted neurostimulation (VR+tSTIM). We compared the effects of this intervention with an active control in 18 participants with chronic neuropathic pain over multiple days. VR+tSTIM led to a clinically significant reduction in self-reported pain intensity. This reduction was accompanied by sensory measures, with participants in the VR+tSTIM group showing enhanced tactile acuity and improved proprioceptive accuracy, effects that did not appear in the control group. Treatment effectiveness was further associated with cortical EEG signatures of decreased gamma and delta power together with increased alpha power. These findings identify potential sensory and cortical biomarkers associated with analgesia and suggest that pain relief in neuropathy may involve the modulation of both peripheral and central mechanisms. This comprehensive assessment paradigm establishes a foundation for the objective monitoring of treatment efficacy and advances the search for mechanistic biomarkers of pain modulation in clinical neuroengineering. This study was approved by the Kantonale Ethikkommission Z&#xfc;rich (Nr. 2021-02258).

Gait and cognitive dysfunction in people with stroke are often exacerbated during dual tasks, limiting their ability to perform daily tasks. This study aimed to investigate the effects of anodal transcranial direct current stimulation (tDCS) over the left dorsolateral prefrontal cortex (DLPFC) on dual-task performance in individuals with chronic stroke. Unlike previous studies that primarily targeted motor cortical regions (e.g., M1), our innovative approach focuses on stimulating the DLPFC, a key cognitive region, to enhance cognitive-motor resource allocation, reduce dual-task costs (DTCs), and ultimately improve both mobility and cognitive performance. In this sham-controlled, within-subject crossover study, 53 participants with chronic stroke underwent two randomized tDCS sessions (active and sham) over 1 week. Following 30&#xa0;min of stimulation, a comprehensive assessment was conducted to evaluate both motor and cognitive DTCs, gait parameters, cognitive functions across various single- and dual-task conditions. Anodal tDCS over the left DLPFC significantly improved lower limb mobility, including gait speed (p&#x2009;<&#x2009;0.001), stride length (p&#x2009;<&#x2009;0.001), and foot strike angle (p&#x2009;<&#x2009;0.001), compared to sham stimulation. Enhanced cognitive performance was observed during dual-task conditions, with significant improvements in Random Number (p&#x2009;=&#x2009;0.001), Counting Forward (p&#x2009;<&#x2009;0.001), and Serial Subtraction (p&#x2009;=&#x2009;0.001) tasks. Notably, tDCS reduced DTCs by improving cognitive response rate during Serial Subtraction (p&#x2009;=&#x2009;0.002). These effects were specific to dual-task conditions, as no significant changes were observed during single-task conditions. Anodal tDCS over the left DLPFC reduces DTCs by enhancing cognitive-motor integration, leading to significant improvements in both mobility and cognitive efficiency in individuals with chronic stroke. This novel approach highlights the DLPFC as a promising therapeutic target for addressing dual-task impairments in stroke rehabilitation. Trial registration: This trial was registered at ClinicalTrials.gov (NCT06818240). The online version contains supplementary material available at 10.1186/s12984-026-01890-2.