ABSTRACT Upper‐limb amputation disrupts natural somatosensory pathways and impairs motor control, increasing reliance on visual feedback during prosthesis operation and reducing intuitiveness and embodiment. Although haptic feedback can improve controllability, most commercial myoelectric prostheses still lack effective closed‐loop feedback. Here, we present a flexible Bidirectional Haptic Feedback System (BHFS) integrating triboelectric multidimensional tactile sensing with multichannel transcutaneous electrical stimulation (TES). The customized Triboelectric Flexible Tactile Sensor (TFT‐Sensor), featuring a strontium titanate (SrTiO 3 )‐modified triboelectric layer with a pyramid microstructure, achieves a 75% increase in open‐circuit voltage compared to the baseline, enabling zero‐power sensing of pressure, shear force, and surface texture. A flexible electrode armband driven by a multichannel electrical stimulator delivers spatiotemporally encoded electrical patterns through closed‐loop mapping algorithms, providing intuitive multidimensional somatosensory feedback. In simulated prosthetic‐hand experiments under audiovisual deprivation, subjects successfully adjusted grip force in real time, maintained stable grasping, and distinguished different surface textures. By establishing a closed‐loop “sensing‐mapping‐stimulation” framework, the proposed system offers a promising strategy for restoring naturalistic somatosensation in prosthetic devices and advancing rehabilitation engineering and human–machine interfaces.

Scalable fabrication of conductive silk textiles for integrated motion monitoring and adaptive thermal management.

Mapping knowledge structure and emerging trends in non-invasive brain-computer interface for stroke rehabilitation.

Multiscale spatiotemporal neural network with multi-attention mechanism using brain partitioning for motor imagery recognition.

Stroke remains a leading cause of long-term disability worldwide, and rehabilitation is essential for recovery. Although artificial intelligence (AI)-related technologies have received growing attention in stroke rehabilitation, the knowledge structure and thematic evolution of this interdisciplinary field remain unclear. To conduct a bibliometric analysis of AI-related research in stroke rehabilitation from 2005 to 2024 and map publication trends, major contributors, thematic clusters, and emerging topics. Relevant publications were retrieved from the Web of Science Core Collection (WoSCC), including SCI-Expanded and SSCI, on November 30, 2024. Only English-language articles and review articles published between January 1, 2005, and November 30, 2024 were included. A total of 3436 records were analyzed using CiteSpace 6.4.R1 Basic, GraphPad Prism 10.1.2, and biblioshiny in R. Analyses covered publication trends, collaboration networks, journal distribution, keyword co-occurrence, clustering, and burst detection. Publication output increased markedly over time, with the United States contributing the largest number of publications. The Swiss Federal Institutes of Technology Domain was among the leading institutions, and Rocco Salvatore Calabrò was among the most productive and highly cited authors. Core publication venues included the Journal of NeuroEngineering and Rehabilitation and IEEE Transactions on Neural Systems and Rehabilitation Engineering. The literature mainly focused on virtual reality, upper-limb rehabilitation, rehabilitation robotics, machine learning, cognitive rehabilitation, and transcranial direct current stimulation. Recent burst terms, including machine learning, artificial intelligence, and deep learning, indicated growing attention to data-driven rehabilitation approaches. AI-related research in stroke rehabilitation has expanded substantially, with increasing emphasis on adaptive, data-driven, and technology-assisted approaches. This study provides a descriptive overview of the field's major trajectories, emerging gaps, and interdisciplinary directions, and may help inform future research and translational exploration.

Objective: This pilot study explored the feasibility of implementing 3D-printed assistive devices in a hospital-based stroke rehabilitation setting. Methods: Feasibility was evaluated across four domains: acceptability, demand, implementation, and limited efficacy testing. Data were collected from both users and providers. Stroke survivors in the intervention group (n = 15) received a 3D-printed assistive technology intervention emphasising user-centered design and user-provider collaboration. Historical controls (n = 31) receiving usual care were identified from medical records. Propensity score matching generated nine matched pairs for comparison. User outcomes included the Functional Independence Measure (FIM) and the Vitality Index (VI), while occupational therapists’ perspectives (n = 10) as providers were assessed using the Japanese version of the modified Technology Acceptance Model questionnaire for 3D-printing technology (TAM-J). Results: Good acceptability was demonstrated, as all stroke survivors in the intervention group consistently used 3D-printed assistive devices in daily activities, and occupational therapists reported positive technology acceptance on the TAM-J. Strong demand was observed among stroke survivors with moderate to severe upper-extremity impairment. Regarding implementation, there were no dropouts, and user-centered devices were adopted through user-provider collaboration. In limited efficacy testing, no additional improvements in the FIM and VI scores were observed compared with controls. However, the intervention helped stroke survivors address their daily challenges. Conclusion: These findings suggested that integrating 3D-printed assistive devices into clinical workflows could be feasible. Future research needs to employ sensitive, user-centered outcome measures and collaborate with designers or rehabilitation engineers to improve the efficiency and quality of device development.

Background Autism spectrum disorder (ASD) is characterized by persistent difficulties in social communication and interaction, alongside restrictive and repetitive patterns of behavior. Virtual reality (VR) has emerged as a promising tool to enhance participation and engagement in individuals with ASD. Objective This bibliometric study aims to map the current research landscape on the intersection between ASD and VR, identifying trends in authorship, publication, geographic distribution, thematic focus, and keyword evolution. Methods A bibliometric analysis was conducted using the Web of Science Core Collection. Classic bibliometric laws and indicators (e.g., Bradford’s, Lotka’s, and Zipf’s laws) were applied to analyze publication volume, citation impact, author productivity, and thematic clustering. Results A total of 398 publications were identified between 2007 and 2024, showing an exponential growth trend (R2 = 0.97). The United States led in total output. Sarkar and Warren were among the most productive authors. The Journal of Autism and Developmental Disorders and IEEE Transactions on Neural Systems and Rehabilitation Engineering published the highest number of relevant articles. Fifty-six articles received 58 or more citations. Thematic clusters revealed emphasis on social skills training, immersive environments, and technological applications in ASD interventions. Conclusion Research on VR applications in ASD has increased exponentially, reflecting growing scientific and clinical interest. These findings provide a foundation for future interdisciplinary investigations and intervention development in developmental disabilities research.

Patients undergoing anterior cruciate ligament reconstruction (ACLR) are at high risk of osteoarthritis or secondary injuries, with abnormal knee contact forces (KCFs) identified as a key factor in joint degeneration. Traditional KCF assessment relies on expensive lab systems while advances in computer vision and AI now enable low-cost alternatives. However, currently available methods oversimplify knee mechanics and neglect compensatory movements, highlighting the urgent need for intelligent, real-time monitoring tools for personalized rehabilitation. Therefore, the aim of this study was to develop and validate an integrated, non-invasive framework for accurate KCFs prediction in ACLR patients during daily activities. We hypothesized that combining enhanced musculoskeletal modeling with a deep learning architecture incorporating spatiotemporal attention would improve the prediction accuracy across multiple movement tasks. This study simultaneously recorded three daily movements of 29 post-ACLR patients using both Vicon and OpenCap. Motion trajectories captured by Vicon were imported into OpenSim for musculoskeletal modeling and KCFs calculation. Dataset comprising OpenCap-derived kinematics and OpenSim-computed KCFs was used to train 3 learning models for the prediction of KCFs in ACLR patients across different movements. Among three models, CNN-BiGRU-Attention model demonstrated the best predictive performance across all three movement tasks (R2walking = 0.973 ± 0.003, R2running = 0.982 ± 0.004, R2descending stairs = 0.951 ± 0.007). CNN and self-attention mechanism collectively enhanced the model's ability to capture key features in ACLR patients' movement data, thereby improving KCF prediction accuracy. Furthermore, for the three daily activities, all models showed superior KCFs prediction performance in running and stair-descent tasks compared to walking. The developed framework successfully achieved high-precision prediction of KCFs. This technological breakthrough not only provides a real-time quantitative tool for rehabilitation monitoring in patients with ACLR, but also facilitates a paradigm shift from static laboratory analysis to dynamic real-time monitoring, with broad application prospects in sports medicine, rehabilitation engineering.

Background/Objectives: Quasi-static inverse dynamics is widely used in biomechanical analyses due to its computational simplicity; however, neglecting inertial effects may introduce joint-specific torque estimation errors during dynamic movements. The purpose of this study was to quantify torque estimation errors introduced by quasi-static assumptions during bodyweight squats performed at different movement frequencies. Methods: A planar MATLAB-based (version R2022a) musculoskeletal model incorporating standard anthropometric parameters was developed to simulate squat motions at 1.00, 0.75, 0.50, and 0.25 Hz. Joint torques calculated using quasi-static inverse dynamics were compared with fully dynamic inverse dynamics at the ankle, knee, and hip. Model agreement was evaluated using Root Mean Square Error (RMSE), normalized percentage error relative to peak dynamic torque, and bootstrapped 95% confidence intervals (CI). Results: Quasi-static modeling produced negligible torque estimation errors at the ankle and knee across all movement frequencies, with percentage errors consistently below 0.1% and narrow confidence intervals. In contrast, the hip joint demonstrated a clear frequency-dependent underestimation of torque when inertial effects were neglected. At 1.00 Hz, the hip RMSE reached 14.4 Nm, corresponding to 14.01% of peak dynamic torque (95% CI: 13.97–14.06%). Error magnitude increased systematically with movement speed. Conclusions: The validity of quasi-static inverse dynamics strongly depends on joint location and movement frequency. While quasi-static models are appropriate for ankle and knee torque estimation during moderate-speed squats, accurate hip torque assessment during faster squats requires full dynamic modeling. These findings provide quantitative benchmarks to inform model selection in biomechanical research, rehabilitation engineering, and assistive device design.

The paper considers the features of developing computational models of prosthetic feet for the human lower limb. The designs of prosthetic foot modules are analyzed. The choice of boundary conditions, material properties, and loading parameters that significantly affect the adequacy and reliability of simulation results is substantiated. The obtained results confirm the feasibility of using biomechanically justified models for performing numerical experiments aimed at improving the functionality, reliability, and comfort of prosthetic feet, as well as optimizing their structural solutions in modern orthopedics and rehabilitation engineering. Figs.: 3 Ref.: 16 items.

ABSTRACT With the increasing global aging population, gerontechnology has emerged as a key approach to improving older adults’ health and quality of life. Spanning rehabilitation engineering, intelligent interaction, digital health, and medical services, the field has expanded rapidly over the past decade. Yet prior work has been limited in articulating the field’s knowledge structure and its evolving trends. To address this gap, this study conducts a bibliometric analysis using the Web of Science Core Collection as the analytical dataset. We integrate CiteSpace, VOSviewer, Scimago Graphica, and Pajek to examine research structure and collaboration networks, and cross-validate keyword signals across Scopus, MEDLINE, and CNKI to enhance sensitivity to emerging topics. In addition, TF-IDF and topic modeling are used to refine keyword extraction and cluster labels, constructing a multidimensional knowledge map of gerontechnology. The findings indicate: (1) a shift of focus from ‘rehabilitation engineering and health management’ toward ‘intelligent ecosystems and data-driven health,’ extending to digital literacy, artificial intelligence, virtual reality, and smart homes – suggesting a movement from compensatory assistance to more preventive and predictive applications; (2) in international collaboration, the United States, the United Kingdom, and China occupy central positions, forming a ‘Western-dominated, Asia-Pacific-rising’ pattern; and (3) since 2020, frontier keywords point to rapid growth in AI-enabled health interventions, cross-scenario applications, and psychosocial factors. Overall, this field-level overview complements traditional narrative and systematic reviews across structure, semantics, and evolution, and offers evidence-informed guidance for researchers and educators designing programs for older adults.

A flexible optical fiber sensor assisted by machine learning was developed and applied on the recognition of hand static poses associated with the Brazilian language alphabet. The sensor stands out for its ease of manufacturing and interrogation, and may help to improve communication between speakers and non-speakers of sign language. The macrobends sensor comprises a single segment of a wave-structured optical fiber encapsulated in a flat hand shape made of silicone. Macrobends distributed along the fingers are simultaneously monitored through the transmission spectrum in the 400–800 nm range. Modifications in the geometry of the flat hand change the guiding conditions and consequently the transmitted light spectrum. Predictive models estimate hand gestures based on pose-dependent spectral features of the captured spectrum. A dataset containing 4000 transmission spectra (100 per pose $$\times $$ 10 poses $$\times $$ 4 days) was used to train and test the models. Tests under repeatability conditions provided a maximum standard deviation of 4.7% in transmittance, reflecting the sensor’s ability to reproduce the same optical signal. The gestures associated with 10 letters of the Brazilian sign language alphabet were recognized by the Support vector classifier model with an accuracy of 97%. This work demonstrates the feasibility of the flexible sensor based on optical fiber macrobends for manual gesture recognition. The ease of installation in gloves and robotic hands makes the sensor a promising tool for application in areas such as sign language communication, rehabilitation engineering, and robotics.

This study aims to develop a prosthetic hand grip control system based on machine vision to improve the quality of life and self-care capacity of patients. In medicine and rehabilitation engineering, machine vision technology has been widely used to design intelligent prostheses to help patients restore limb function. Grip strength control is one of the key challenges in developing prosthetic hands; for example, patients need to appropriately control the grip strength of the prosthetic hand depending on the nature and size of the object to be gripped to prevent it from slipping or being damaged. This study combines machine learning and deep learning techniques to determine object grip information by analyzing images of such objects, including their type, texture, and size, so as to select the appropriate grip strength threshold. The electromyographic gesture-control mode is integrated with the visual recognition system to achieve active detection and control of the intelligent prosthetic hand. This research is also transplanted into the K210 main control board for offline recognition to achieve more efficient real-time performance. The experimental results demonstrate that the system achieves an object recognition accuracy rate of 90%, and the real-machine recognition rate is above 85%. The system successfully implements adaptive grasping for eggs (fragile items) and water bottles (rigid objects).

Electromyography (EMG) signal analysis is essential for evaluating neuromuscular activity, with applications spanning prosthetic control, rehabilitation engineering, clinical diagnostics, and human-computer interfaces. Robust feature extraction transforms raw EMG signals into discriminative representations for pattern recognition and decision-making. This review comprehensively surveys classical and artificial intelligence (AI)-based EMG feature extraction techniques. Traditional methods fall into time-domain, frequency-domain, and time-frequency-domain categories, each balancing computational efficiency and interpretability. Time-domain features—such as mean absolute value (MAV), root mean square (RMS), and zero crossings (ZC) — are simple and well-suited to real-time processing. Frequency-domain descriptors, including power spectral density (PSD), mean frequency (MNF), and median frequency (MDF), reveal insights into muscle fatigue and contraction dynamics. Time-frequency approaches like the short-time Fourier transform (STFT) and the wavelet transform (WT) effectively capture the non-stationary characteristics of EMG signals. Emerging machine learning and deep learning paradigms enable automated feature discovery and superior classification performance. We delineate the strengths, limitations, and context-specific efficacy of these methods, underscoring the potential of hybrid and AI-driven strategies to advance biomedical signal processing.

This dataset addresses the need for multimodal biomechanical recordings during over-ground walking, ramps, and stairs by synchronously capturing electromyographic (EMG), inertial (IMU), and plantar pressure data. We collected data from 45 adults (15 with unilateral transtibial amputation and 30 without amputation) who completed five standardized locomotor tasks: level walking, ramp ascent/descent, and stair ascent/descent. Each participant performed 50 supervised trials. Wireless EMG and IMU sensors (Delsys Trigno Avanti) measured muscle activation and kinematics, while intelligent insoles (XSENSOR) captured plantar pressure distribution. Raw data were saved in.hpf (EMG/IMU) and.XSN (pressure) formats, with processed outputs in.csv files. All data are organized by task and sensor type, including complete participant metadata. Key dataset outputs include time-normalized EMG amplitudes, segment kinematics, and pressure maps across terrains and populations. The dataset was validated technically and experimentally during the acquisition. This resource enables quantitative analysis of gait adaptation and supports machine learning for locomotion classification. Data are provided in accessible formats to foster reuse in biomechanics, rehabilitation engineering, robotics, and clinical gait research.

Abstract The landscape of telehealth has changed drastically over the last decade, extending beyond the remote delivery of clinical services (i.e., telemedicine). Worldwide, there has been continued growth in the popularity of mHealth (mobile health) applications to support physical, mental, and social health. From virtual exercise classes to remote health monitoring systems, mHealth can utilize a variety of mobile technologies such as smart phones, laptops, and wearable devices. mHealth applications can be especially beneficial for people aging with mobility disabilities who experience transportation and accessibility barriers with in-person programs. This session will highlight innovative mHealth solutions from the Rehabilitation Engineering Research Center on Technologies to Support Aging among People with Long-Term Disabilities (RERC TechSAge). Mitzner et al. will present results from the TechSAge Tele Tai Chi clinical trial – an evidence-based tai chi program delivered via Zoom designed to be socially engaging and inclusive of older adults with mobility limitations. Rice et al. will describe the development of a falls detection and monitoring system for older adults who use wheelchairs that integrates smart watch monitoring and a user-facing app. Hsieh et al., will highlight the design a fall risk mHealth app for people with Multiple sclerosis (MS) that enables users to measure and evaluate their fall risk and engage in preventative strategies. Eric Levitan, CEO & Founder of Vivo, an online, live fitness program designed for older adults, will serve as the discussant, engaging participants in interactive discussion and offering an industry perspective on challenges implementing mHealth applications for older adults.

The convergence of wearable technologies and digital twin (DT) systems is transforming rehabilitation engineering, enabling continuous monitoring, personalized therapeutic interventions, and predictive modeling of patient recovery pathways. This review examines the growing role of machine learning (ML) in the development and integration of DTs frameworks in rehabilitation, with a focus on wearable sensor data, security and privacy, edge computing architectures, federated learning paradigms, and generative artificial intelligence (GenAI) applications. We first analyze data collection processes, emphasizing multimodal sensing, signal processing, and real-time synchronization between physical and virtual patient models. We then discuss key challenges related to data security, encryption, and privacy protection, especially in distributed clinical environments. The review then assesses the role of edge computing in reducing latency, improving energy efficiency, and enabling real-time local intelligence feedback in wearable devices. Federated learning approaches are discussed as promising strategies for jointly training ML models without compromising sensitive medical data. Finally, we present new GenAI techniques for generating synthetic data, personalizing digital twins, and simulating rehabilitation scenarios. By mapping current progress and identifying research gaps, this article provides a unified view that connects electronic and biomedical engineering with intelligent, secure, and adaptive DT ecosystems for next-generation rehabilitation solutions. Wearable devices with ML and DTs for rehabilitation are developing rapidly, but their current effectiveness still depends on consistent, high-quality data streams and robust clinical validation. The most promising convergence involves combining edge intelligence with federated learning to enable real-time personalization while preserving patient privacy. GenAI further enhances these systems by simulating patient-specific scenarios, accelerating model adaptation, and treatment planning. Key challenges remain related to standardizing data formats, ensuring comprehensive security, and seamlessly integrating these technologies into clinical processes.

Abstract Water infiltration areas are places where rainwater seeps into the soil and becomes groundwater through infiltration. Poor water absorption can cause an increase in the volume of surface water flow; therefore, it is necessary to identify water catchment areas that play an important role in maintaining the environment and stability of the water cycle using a Geographic Information System model. The Segeri Watershed is located in Pangkajene and Kepulauan Regency and Barru Regency, where land conversion occurs frequently, causing frequent flooding. This study aimed to identify the condition of the water catchment area in the Segeri Watershed and analyze the factors that most influence the condition of the water catchment area in the Segeri Watershed. This study used the overlay method to assess potential infiltration parameters and actual infiltration based on the Procedures for Compiling Forest and Watershed Land Rehabilitation Engineering Plans (RTkRHL-DAS,2009). Data were collected through spatial data processing, soil sampling, and land cover ground checks. The results of the study showed that the water infiltration area in th Segeri Watershed has four conditions: of the total area of the Segeri Watershed, which is 18,933.62 ha, approximately 44.25% has good conditions. The primary factor influencing the condition of the water absorption area in the Segeri Watershed is land cover; specifically, areas dominated by good conditions exhibit greater actual infiltration, particularly in regions covered by secondary dryland forest, which constitutes 20.24% of the total water catchment area

This paper presents the results of an advanced inspection, structural assessment, and development of rehabilitation and strengthening strategies for Concrete Bridge No. 44, located on National Road N2, Prishtina - Blacë. The primary objective of this study is to identify effective methods for restoring load-bearing capacity and improving the operational safety of the bridge, in compliance with contemporary design standards as prescribed by the Eurocodes, considering the current degradation state of its structural elements and the critical importance of this infrastructure within the national road network. The assessment methodology involved systematic visual inspections, non-destructive testing, and numerically modeled structural analyses to evaluate the existing performance of the main load-bearing beams, columns, and other key components. Based on the collected data, modular and phased intervention proposals have been developed, allowing for the prioritized implementation of repair measures according to the severity of identified deficiencies and the functional priorities of the bridge. The results of this study provide both practical and scientific contributions to the planning of interventions on existing structures built in previous decades, combining modern rehabilitation engineering approaches with international standards of structural design and safety control.

This study systematically develops a comprehensive framework for human-centered prosthetic design, directly responding to the evolving needs of domestic prosthesis users. Integrating theoretical modeling with empirical investigation, the research identifies core design determinants aligned with user expectations. Employing a mixed-method approach, including large-scale surveys and deep interviews, the study explores Chinese prosthetic users’ functional, emotional, and social integration needs. Findings reveal a notable gap between current products and user aspirations, especially in biomechanical adaptability, aesthetic customization, intelligent interaction, and seamless daily integration. A novel ‘cultural context–social integration’ design paradigm is proposed, transcending purely technical models. These interdisciplinary insights support progress in rehabilitation engineering, assistive technology, and inclusive product policy. The system establishes a comprehensive framework for human-centered prosthetic design, directly addressing the evolving needs of domestic prosthetic users. Through integrating theoretical modeling with empirical research, it identifies core design elements that align with user expectations.