Magnetic guidewires are increasingly investigated for vascular intervention, yet the device-level impact of magnetic actuation on delivery time and proximal pushing force remains insufficiently quantified. In this work, MANUS (Maneuverability Assessment of magNetic gUidewireS with proximal delivery force) is proposed as a quantitative framework for evaluating magnetic guidewires in a complex, multi-branch vascular phantom. Two magnetic guidewires are compared with a passive guidewire while six novice operators navigate three sequential targets using a hand-held external magnet for distal steering. Across operators, navigation time is shortened and proximal pushing force is reduced when magnetic guidewires are used. Among the tested devices, one magnetic guidewire achieves the best overall performance, reducing mean delivery time by 28.5% and significantly decreasing pushing-force metrics relative to the passive guidewire (p < 0.001). These results suggest that magnetic actuation can improve navigation efficiency while lowering proximal loading, potentially reducing operator workload and supporting safer, faster endovascular navigation.

Predicting knee joint trajectories from surface electromyography (sEMG) signals holds a significant application value in various fields such as rehabilitation engineering and prosthetics control. However, existing prediction methods often struggle to achieve satisfactory performance due to limited dataset sizes and poor cross-subject generalization capabilities. In this paper, we propose an effective framework that integrates motion decoupling with a conditional diffusion model to address these challenges. Our approach decomposes knee joint angles into shared motion patterns across subjects and individual-specific amplitude parameters, enabling dual-task collaborative modeling that considers both commonalities and individual differences. Furthermore, the conditional diffusion model is employed to generate high-quality synthetic sEMG samples, effectively expanding the available data resources. Experiments conducted on data from 11 subjects demonstrate that our approach achieves a Root Mean Square Error ( RMSE ) of 3.59 ± 0.88 ^∘ , outperforming the non-decoupled model (4.61 ± 1.58 ^∘ ), the model without diffusion (4.85 ± 1.62 ^∘ ), the Bidirectional Long Short-Term Memory (Bi-LSTM) (6.75 ± 1.33 ^∘ ) and the traditional LSTM baseline (6.88 ± 1.59 ^∘ ).

Abnormal gait recognition plays a critical role in healthcare, particularly for the early diagnosis and continuous monitoring of neurological and musculoskeletal disorders, such as Parkinson’s disease and orthopedic injuries. This study proposes MSCAF-Gait, a Multi-Scale Cross-Attention Fusion Network designed specifically for abnormal gait recognition using foot pressure sensors. MSCAF-Gait incorporates multi-scale convolutional modules with channel and spatial attention mechanisms to effectively capture features across temporal, channel, and spatial dimensions. A novel cross-attention fusion module further enhances feature representation, enabling precise recognition of diverse abnormal gait patterns. To facilitate this research, we introduce the Pressure-Insole Abnormal Gait (PIAG) dataset, comprising gait data associated with common neurological and musculoskeletal abnormalities. Extensive experiments on the publicly available Gait in Parkinson’s Disease (GaitinPD) dataset and our self-constructed PIAG dataset validate the effectiveness of MSCAF-Gait. Specifically, the model achieves 99.61% accuracy in Parkinsonian gait recognition and 98.88% accuracy in Parkinson’s severity classification. On the PIAG dataset, which includes multiple abnormal gait patterns, MSCAF-Gait attains a high accuracy of 99.42%. Notably, these results are obtained with a lightweight architecture characterized by reduced FLOPs and parameter count, demonstrating that MSCAF-Gait offers both high accuracy and computational efficiency, making it well-suited for real-time deployment on wearable platforms.

In injury-induced mortality among the elderly, falls represent one of the primary contributing factors. Thus, assessing fall-related risks holds critical significance for elderly health management. Existing fall risk assessments depend on instruments, clinical judgment, and environmental checks, but are often subjective and inconsistent. This study develops Bino-GaitRisk-Transformer, a novel binocular vision + Transformer architecture for fall risk prediction. By leveraging a confidence-weighted 3D reconstruction algorithm, it achieves millimeter-level joint localization accuracy (mean error <10 mm), reducing error by 40 >20

With the rapid development of Rehabilitation Lower Extremity Robotic Exoskeletons (RLEEX) technology, significant advancements have been made in Human-Robot Interaction (HRI) methods. These include traditional physical HRI methods that are easily recognizable and various bio-electrical signal-based HRI methods that can visualize and predict actions. However, most of these HRI methods are contact-based, facing challenges such as operational complexity, sensitivity to interference, risks associated with implantable devices, and, most importantly, limitations in comfort. These challenges render the interaction less intuitive and natural, which can negatively impact patient motivation for rehabilitation. To address these issues, this paper proposes a novel non-contact gesture interaction control method for RLEEX, based on RGB monocular camera depth estimation. This method integrates three key steps: detecting keypoints, recognizing gestures, and assessing distance, thereby applying gesture information and augmented reality triggering technology to control gait movements of RLEEX. Results indicate that this approach provides a feasible solution to the problems of poor comfort, low reliability, and high latency in HRI for RLEEX platforms. Specifically, it achieves a gesture-controlled exoskeleton motion accuracy of 94.11% and an average system response time of 0.615 seconds through non-contact HRI. The proposed non-contact HRI method represents a pioneering advancement in control interactions for RLEEX, paving the way for further exploration and development in this field.

Falls are common among the elderly and can lead to physical injuries, psychological fear, and social isolation, all of which severely affect their quality of life. Therefore, accurately assessing fall risk in older adults is crucial for timely preventive measures to reduce fall incidents and related injuries. This study proposes a novel method for predicting fall risk in the elderly based on Kalman filtering and binocular vision for 3D motion reconstruction combined with a CNN-LSTM-Attention model. The model includes a feature extraction layer that uses MediaPipe to identify keypoint coordinates, while Kalman filtering is applied to reduce noise caused by endpoint jitter and irrelevant movements. Based on the reconstructed 3D data, motion features related to fall risk, such as gait speed and stride length, are calculated. The CNN-LSTM-Attention model is then employed to capture spatial and temporal features, with the attention mechanism enabling the network to dynamically adjust the weight of each time step by modeling global dependencies during training. Experimental validation of the proposed evaluation system demonstrates a model accuracy of 0.831 and an Fl-score of 0.813, providing a comprehensive and reliable assessment for fall prevention in the elderly. This system can be widely applied in settings such as nursing homes, home monitoring, and hospital rehabilitation departments. By accurately monitoring the motion status of the elderly and assessing their fall risk, it enables early warning and timely intervention, thereby effectively reducing the incidence of falls and enhancing the quality of life and safety of elderly individuals.

Constrained environments impact the motion dynamics of magnetic guidewires, introducing uncertainties into control. To address this challenge, a robust control method for magnetic guidewire navigation is proposed. First, a kinematic model is formed to characterize the behavior of the magnetic guidewire, which integrates both steering and advancement motions. Based on this model, a direct adaptive control (DAC) method is proposed, where a learnable parameter is employed to dynamically compensate for environmental disturbances, thereby significantly improving trajectory tracking performance. Additionally, Lyapunov stability analysis is further employed to verify the stability of the proposed control method. Finally, extensive simulation experiments demonstrate its effectiveness, achieving a Root Mean Square Error (RMSE) of 0.0182 m in tracking a desired trajectory. These results highlight the robustness and practical potential of the proposed framework for vascular intervention tasks.

Objective: Aiming to quantify and analyze disease-induced alterations in human movement, we explored the co-joint synergy patterns in locomotion through a vision-based co-joint synergistic attention algorithm. Methods: We recruited 30 participants (including 15 poststroke patients and 15 healthy individuals) and extracted their 3D visual motor data for the joint feature coupling by a serial attention module. And we designed a dual-stream classification module for preclassification based on the spatio-temporal characteristics of the data. Then we extracted the important co-joint synergy patterns by a looping mask module and the co-joint synergy variability score. Results: Through the co-joint synergistic attention algorithm, we found significant differences in joint synergy patterns between post-stroke patients and healthy individuals during upper and lower limb tasks. Furthermore, we obtained quantitative results on the effect of specific diseases on cojoint synergy patterns among healthy individuals and patients. The validity of the result was verified by comparing with the commonly used Non-negative Matrix Factorization (NMF) and the Muscle Synergy Fractionation (MSF) methods. Conclusion: Specific diseases can cause changes in human movement patterns, and by the co-joint synergistic attention algorithm we can analyze the alterations in joint synergies and also quantify the importance of different synergy groups. Significance: This research proposes a new approach for identifying specific co-joint synergy patterns arising from disease-altered biomechanics, which provides a new targeted protocol for the rehabilitation process.

With the continuous concern for the disabled and the elderly, intelligent prosthetics and service robots have been widely applied. This paper provides a method for gesture recognition using forearm surface electromyography (sEMG), including an adaptive channel selection method to simplify the sEMG measurement. Based on the forearm muscle groups corresponding to different movements, surface skin areas are divided, and the Myo bracelet is used to collect sEMG signals from these areas. A method combined with channel attention module, multi-channel relationship feature extraction module and multi-scale skip connection module is built to adaptively select the signals from certain skin areas and recognize the seven gestures during experiment. The comparative experimental results indicate that this method can adaptively extract the optimal channel combination and show effective recognition results. It improved the practicability for the sEMG-based gesture recognition.

Dear Editor, This letter investigates the system development of a multi-joint rehabilitation exoskeleton,and highlights the subject-adaptive con-trol factors for efficient motor learning. In order to enable the natural mobility of the human upper extremity,we design the shoulder mechanism by arranging three rotational joints with acute angles,and adopt a serial chain structure for the fully constructed system.

This study addresses the challenge of designing finite-time output feedback control for a category of 5 -degree-of-freedom upper-limb exoskeleton robotic systems. This investigation introduces an observer which is devised to ensure the convergence of observer errors to a small neighborhood of zero within finite time. Then, a finite-time output feedback controller is subsequently developed, utilizing the designed finite-time observer and the sliding model control method. The application of the homogeneous Lyapunov function, in conjunction with the Lyapunov method, demonstrates that the tracking error can be confined to a tight region around zero within a finite time frame, provided that only output measurement is utilized. Ultimately, an illustrative case is presented to demonstrate the efficacy of the introduced control strategy.

This paper addresses the predefined-time control problem for upper limb rehabilitation robots. Utilizing the time-varying function technique and sliding mode control, a predefined-time control scheme is developed. The proposed control law is continuous and non-singular, ensuring that the tracking error converges to a small bounded region within the predefined time. Finally, a numerical example is provided to validate the effectiveness of the proposed control scheme.

Current epileptic detection methods primarily utilize machine learning, wavelet transforms, and multi-feature fusion, yet often inadequately distinguish between interictal and ictal status. To address the limitation, this study proposes a novel enhanced EEG decoding framework that systematically integrates time-domain, frequency-domain, and nonlinear characteristics of epileptic signals. Our method incorporates three key innovations: (1) comprehensive multi-domain feature fusion architecture, (2) multi-head attention mechanism enabling adaptive feature weighting, and (3) multi-task learning architecture for classification of healthy, interictal, and ictal status. Experiments on EEG data from five healthy subjects and five epileptic patients demonstrated robust performance with 94.00

With the increasing importance of the exoskeleton robot in motion assistance and rehabilitation therapy, its precise control is particularly important. In this paper, an efficient control strategy combining the time-varying function, the predefined-time observer, and the backstopping method is proposed to enhance the performance steadiness and speed of the exoskeleton robot in a complex dynamic environment. By designing a time-varying function combined with a predefined-time observer for precisely evaluating the system's state, the rapid convergence of the system state to a desired state is realized in this study in the predefined time without requirement of full-state measurements. Through numerical simulation examples, the control system's effectiveness undergoes a thorough assessment. The results show that the prosed control scheme can enforce the tracking error to a small region near zero within the prescribed duration of time, and provide the rehabilitation robot with good anti-interference ability and high control accuracy.

This study investigated age-related characteristics using multi-channel surface electromyography (sEMG) signals. Forearm sEMG signals were captured from 39 participants (aged 60-80 years) during nine standardized daily tasks. Unsupervised clustering of sEMG-derived features identified biomarkers that classified subjects into three aging stages: Stage I (mostly 60-69 years), Stage II (mostly 70-74 years), and Stage III (mostly >= 75 years). Clinician assessments confirmed accelerated motor aging in some participants, categorizing them into stages exceeding their chronological age. sEMG features associated with neuromuscular decline showed high homogeneity in subjects >= 75 years, consistently classifying them into Stage III. For participants under 75 years, sEMG signal complexity features effectively differentiated Stage I (60-69 years) from Stage II (70-74 years). This study establishes sEMG-derived biomarkers for assessing motor function decline in older adults (60-80 years), thus supporting personalized rehabilitation strategies aligned with motor aging stages. Copyright (c) 2025 The Authors. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)

Freezing of Gait (FOG) is a significant symptom contributing to gait dysfunction in Parkinson’s disease (PD) patients. Most current methods for assessing FOG severity often overlook the interpretability of the extracted gait features. In this study, we design a multimodal gait feature dataset with rich physical significance, including kinematics, kinetics, and spatiotemporal modalities. We also propose a graph-based multimodal fusion framework (GMFF) to accurately quantify FOG severity. GMFF employs the graph attention mechanism to extract modality-specific features and utilizes the generalized canonical correlation analysis (GCCA) algorithm as the core of the feature fusion module. We provide the double-hurdle output module to address the impact of the zero-inflation problem on the performance of GMFF. We evaluate the performance of GMFF on a public PD gait database using five-fold cross-validation. The results demonstrate that GMFF achieves an accuracy of 0.978 in identifying patients with FOG and a root mean square error of 0.449 in quantifying FOG severity. Using the interpretability of GMFF, we identify the gait feature set that effectively characterizes the gait patterns of PD patients and then explore the impact of FOG symptoms on their walking ability under both the “ON” and “OFF” medication states. Thus, this study has the potential to provide valuable insights into the clinical monitoring and management of PD patients.

Gait analysis is significant in clinical and rehabilitation medicine for evaluating lower limb diseases. However, the existing gait measurement and analysis technologies rely on large-scale professional equipment and physicians, and their application scenarios are limited. In this work, we design a lightweight motor function analysis system, including multi-sensor camera and gait extraction - assessment algorithm. The system first uses binocular active infrared sensors to collect information about the subjects and build a body model. Then, the binocular red-green-blue (RGB) camera is used to collect the subject’s motion information, and the established subject model is used to identify and fit the human movement posture and extract the gait features. We build a convolutional neural network to process gait features and evaluate the disease status of subjects. The effectiveness of the system is verified by experiments. We recruit healthy subjects and patients to validate the system and achieve an accuracy of 92.86% on the disease assessment task, verifying the effectiveness of the system.

Rehabilitation robots based on variable stiffness elastic actuators exhibit strong human-robot interaction char-acteristics due to the specific features of the actuators, including flexible drive and variable stiffness output capabilities. However, to ensure the robot can achieve rehabilitation actions smoothly, safely, and precisely, accurate modeling and control of the variable stiffness elastic actuators becomes a burning question. In order to address this challenge, a dynamic model is established considering the actuator's own gravity factor and the influence of disturbances, and a nominal dynamic model is designed. Accurate tracking of output location and stiffness is reached by designing a radial basis function (RBF) network to approximate the actual dynamic model. Through simulation experiments, it is verified that the block approximation RBF network adaptive controller designed in this article has lower actuator tracking error and higher control accuracy than the proportional integral derivative (PID) controller.

Cyclograms are useful for describing interjoint coordination in clinical gait analyses. In this paper, we establish a model for generating hip-knee cyclograms at different walking speeds by using B-Spline curves, which could provide reference standards for evaluating hip-knee joint coordination. We use the shape similarity between individual and reference cyclograms to develop a hip-knee joint coordination assessment metric: the dynamic time warping-shape context (DTW-SC). The DTW algorithm is used to match corresponding points between the hip-knee cyclograms, and the matched results are visualized to show abnormal changes in gait patterns. The distance between corresponding points is determined through shape context descriptors and applied to quantify the hip-knee joint coordination of the subjects. To verify the effectiveness of the proposed approach, a stereo vision-based portable lower limb motion capture system is developed to collect hip and knee joint angle data from five healthy and five hemiplegic subjects walking at comfortable speeds. Experimental results show that the range of the modeling error is [0.84(degrees), 2.76(degrees)]. DTW-SC metric can visualize and quantify the abnormal patterns of the subject's hip-knee joint coordination. The work is meaningful in evaluating the hip-knee joint coordination of hemiplegic patients.

This paper presents the modelling and control of a novel upper-limb exoskeleton with a serial structure. The dynamic modelling is performed by using the Newton-Euler method, followed by linear transformations to obtain a minimal parameter set containing unknown dynamic parameters. By considering uncertain factors such as inertial forces and transmission losses within motor internal reducers, a periodic trajectory is designed for the purpose of parameter identification. The weighted least squares method was employed for this purpose, estimated values for the dynamic parameters are calculated and compared with actual values by employing the root mean square error and mean absolute error. Additionally, a zero-force control based on an outer-loop PID strategy has been devised, and the viability of the dynamic model and the controller has been demonstrated through experimentation. The results demonstrate that the proposed upper-limb exoskeleton can be employed to enhance the efficacy of robot-assisted rehabilitation.