IEEE transactions on medical robotics and bionics最新文献

{"title":"Edge-Based Robot-Assisted Hand Gesture Recognition for Autism Intervention Using Hybrid Keyframe Selection and Contrastive Learning","authors":"Sougatamoy Biswas;Soumyajeet Mahapatra;Anup Nandy;Asim Kumar Naskar","doi":"10.1109/TMRB.2026.3680976","DOIUrl":"https://doi.org/10.1109/TMRB.2026.3680976","url":null,"abstract":"Children with Autism Spectrum Disorder (ASD) often face challenges in expressing and interpreting social gestures essential for social skill development. Traditional approaches depend on intrusive wearable sensors and manual supervision, with minimal adaptation to real-time, resource-constrained environments. To address these limitations, we present a real-time vision-based human–robot interaction (HRI) framework that autonomously supports gesture learning and correction without therapist intervention. The system employs MediaPipe-based skeletal landmark extraction with a novel hybrid keyframe selection strategy. This strategy integrates entropy-based filtering to retain semantically rich frames and a clustering-based mechanism to identify representative frames from similar gesture groups. Spatial features are modeled using one-dimensional convolutions across hand joints to capture intra-frame structural dependencies. The framework executes gesture classification and correctness evaluation simultaneously, with contrastive learning incorporated to enhance the separability of learned embeddings. An in-house assistive robot, named CARAT (Caregiver Autism Robot with AI-based Technology), has been developed to realize this framework. CARAT is equipped with a gesture-performing arm, deployed on a Jetson Orin Nano edge device, and provides real-time verbal feedback alongside physical gesture demonstrations. The dataset for evaluation was collected using CARAT, involving participants from both healthy children and children with ASD. The framework demonstrates strong recognition performance, achieving 98.85% accuracy with healthy participants, 96.42% with autistic participants, and 97.41% on the combined dataset of both groups. These results validate the effectiveness of the proposed approach for hand gesture recognition in edge-deployed assistive robots for autism intervention.","PeriodicalId":73318,"journal":{"name":"IEEE transactions on medical robotics and bionics","volume":"8 2","pages":"811-822"},"PeriodicalIF":4.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148509306","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Autonomous Robotic Liver Ultrasound via Vision-Guided Initialization and Imitation Learning","authors":"Ji Ma;Yue Zhao;Zhuang Liu;Yi Shen;Jianxing Liu","doi":"10.1109/TMRB.2026.3680972","DOIUrl":"https://doi.org/10.1109/TMRB.2026.3680972","url":null,"abstract":"Liver disease is a leading cause of mortality worldwide. Ultrasound (US) examination is a primary modality for monitoring and assessing the risk of liver disease. Robotic ultrasound system (RUSS) has emerged as a promising approach to reduce dependence on sonographers and to improve the standardization and reproducibility of US examination. To advance liver US examination, this study proposes a fully autonomous two-stage methodology for robotic liver US examination, integrating vision-guided initial positioning and imitation learning (IL). In the initial positioning stage, the probe autonomously establishes initial contact with patients using RGB-D images and determines the transition timing to the IL stage based on real-time US images. In the IL stage, scanning skills–represented by robot trajectories in Cartesian space and corresponding US image state trajectories–are learned and reproduced using an enhanced IL method, Adaptive Ensemble Transfer Bayesian Interaction Primitive (AET-BIP). Compared with its original formulation, AET-BIP reduces fluctuations in scanning phase velocity, thereby improving stability. To evaluate repeatability, the robot’s end-of-scan poses were measured under varying initial conditions in the IL stage, yielding position and orientation repeatability errors generally below 2.50 mm and 0.020 rad, respectively. Comprehensive experiments incorporating both stages validate the feasibility and adaptability of the proposed methodology, highlighting its potential for standardized and reliable clinical liver US examination.","PeriodicalId":73318,"journal":{"name":"IEEE transactions on medical robotics and bionics","volume":"8 2","pages":"704-716"},"PeriodicalIF":4.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148507807","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An Ankle-Driven Proprioceptive Rehabilitation Robot During Gait: Design, Control, and Evaluation","authors":"Liziyi Hao;Zhaocheng Zhou;Honghao Zheng;Xiangyu Wang;Jianda Han;Ningbo Yu","doi":"10.1109/TMRB.2026.3680991","DOIUrl":"https://doi.org/10.1109/TMRB.2026.3680991","url":null,"abstract":"Perturbation-based training is essential for improving proprioception and reducing fall risk, yet many robotic platforms emphasize sagittal-plane assistance or apply three-dimensional perturbations via trunk, and configurable multidirectional perturbations directly at the foot-ankle interface during walking are still lacking. In this study, we present a robotic system for proprioceptive training that programmatically delivers plantar perturbations in multiple directions during gait. The system integrates a 10-DoF series-parallel hybrid mechanism, an immersive visual interface, and a suite of task modes to reproduce unstable terrains and generate controlled perturbations. To ensure stable operation, a composite control method combining iterative learning with neural-network disturbance estimation was designed, and reduced the trajectory-tracking errors by over 60% in comparison with PID and sliding mode control. Experiments involving ten healthy participants were conducted under mediolateral perturbations and sEMG-based muscle synergy analysis revealed training-induced neuromuscular adaptations, marked by the progressive activation of key frontal-plane balance muscles - gluteus medius, tensor fasciae latae, and peroneus longus - and a reorganization of task-specific synergies under higher perturbation levels. These findings demonstrate that the elaborated proprioceptive training with controlled perturbations drives neuromuscular adaptation, highlighting the potential of the proposed robotic system as a viable tool for balance rehabilitation.","PeriodicalId":73318,"journal":{"name":"IEEE transactions on medical robotics and bionics","volume":"8 2","pages":"832-843"},"PeriodicalIF":4.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148509311","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Design of a Magnetic Dual-Needle Biopsy Capsule Robot Resilient to Cross-Infection and Mixed Contamination","authors":"Bo Ye;Zuolei Chen;Shufang Wang;Sixian Liu;Bo Wang;Zhi Shu;Yihan Xu;Baoshun Liu;Shoujun Dai;Anh Nguyen;Amit Krishna Dwivedi;Sheng Liu","doi":"10.1109/TMRB.2026.3680996","DOIUrl":"https://doi.org/10.1109/TMRB.2026.3680996","url":null,"abstract":"Capsule robots have demonstrated great potential for comfortable, non-invasive diagnosis of the gastrointestinal tract. With the advancement of research in capsule endoscopy, its functionality has expanded from initial imaging capabilities to include biopsy operations. However, most biopsy capsule robots cannot separately place the tissues obtained during biopsy. The cross-infection caused while using the same biopsy tools at different biopsy sites, and inaccurate results due to mixed contamination of the pathological tissues, still need to be addressed. This paper presents a novel magnetic Dual-needle Biopsy Capsule Robot (DBCR) design to solve this problem. The capsule design was simplified into a double needle switching and anchoring structure in the capsule, modelled based on a retractable pen’s extension and retraction control mechanism. Additionally, the design was equipped with magnets and springs of different sizes to complete the needle switching mechanism and extract the tissue samples. The force of the external control magnet on the magnet inside the capsule was simulated to determine the control distance to achieve different functions and ensure sampling reliability. The prototype was designed, manufactured, and tested using in vitro experiments. The achieved results show that the proposed design of the DBCR is promising and can potentially obtain biopsy samples, preventing cross-infection and mixed contamination.<italic>Note to Practitioners—</i> The proposed DBCR design can potentially address the limitations of current systems in obtaining biopsy samples, mainly by preventing cross-infection and mixed contamination, in double-needle designs. The proposed design offers a dual-needle storage and switching mechanism based on a retractable pen-inspired linear-to-rotational motion conversion system for compartmentalized sampling and the gradient magnetic field control of embedded magnets and springs to ensure needle activation only during biopsy. Thus, the proposed design finds its suitability for use in biopsy.","PeriodicalId":73318,"journal":{"name":"IEEE transactions on medical robotics and bionics","volume":"8 2","pages":"894-905"},"PeriodicalIF":3.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147923934","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improved Classification Accuracy of Convolutional Neural Networks With Connected Layers for Electroencephalogram Data","authors":"Viet Quoc Huynh;Tuan Van Huynh","doi":"10.1109/TMRB.2026.3681027","DOIUrl":"https://doi.org/10.1109/TMRB.2026.3681027","url":null,"abstract":"Advanced deep-learning techniques have been employed to enhance the field of brain-computer interfaces through the processing of electroencephalogram (EEG) signals. Recent studies have shown application of convolutional neural networks (CNNs) with manually extracted EEG features to be a promising approach to improve classification accuracy. However, these models require complex input features to generate output decisions. This study aimed to improve EEG recognition by developing novel CNN architectures based on branch CNNs: the step CNN (the base model) and the input- and layer-additive CNNs, which incorporate connected layers. The models exhibit exceptional performance across established WAY-EEG-GAL benchmark datasets. Additionally, the proposed method was evaluated and optimized through EEG dataset experiments, and surpassed other state-of-the-art techniques. Using connected layers in the models increased the effectiveness in recognizing EEG signals during grab-and-lift classification.","PeriodicalId":73318,"journal":{"name":"IEEE transactions on medical robotics and bionics","volume":"8 2","pages":"918-927"},"PeriodicalIF":3.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147923888","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"IEEE Transactions on Medical Robotics and Bionics Information for Authors","authors":"","doi":"10.1109/TMRB.2026.3684275","DOIUrl":"https://doi.org/10.1109/TMRB.2026.3684275","url":null,"abstract":"","PeriodicalId":73318,"journal":{"name":"IEEE transactions on medical robotics and bionics","volume":"8 2","pages":"C4-C4"},"PeriodicalIF":3.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11519515","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147923894","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"IEEE Transactions on Medical Robotics and Bionics Society Information","authors":"","doi":"10.1109/TMRB.2026.3684273","DOIUrl":"https://doi.org/10.1109/TMRB.2026.3684273","url":null,"abstract":"","PeriodicalId":73318,"journal":{"name":"IEEE transactions on medical robotics and bionics","volume":"8 2","pages":"C3-C3"},"PeriodicalIF":3.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11519499","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147923933","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Configuration Optimization Strategy of Magnetic Soft Robot for Robot-Assisted Intervention","authors":"Siyi Wei;Zhiwei Wu;Jinhui Zhang","doi":"10.1109/TMRB.2026.3680982","DOIUrl":"https://doi.org/10.1109/TMRB.2026.3680982","url":null,"abstract":"Magnetic soft robots (MSRs) are promising tools for precise intravascular interventions due to their flexibility and magnetic actuation capability, yet existing designs face fundamental trade-offs: magnetic continuum robots offer uniform deformation but suffer from weak magnetic responsiveness, while magnet-tip robots provide strong actuation but lack flexibility and workspace. Previous optimization strategies have largely treated geometry, material distribution, and magnetic configuration separately, leaving their coupled effects insufficiently understood. Here, we introduce a comprehensive theoretical model that unifies elastic, magnetic, gravitational, and external forces to analyze the MSR’s dynamic response and energy distribution. Guided by this model, we design a model-guided MSR with tailored magnetic distribution and structural configuration, achieving a normalized forward workspace of 0.28, which is nearly eight times larger than that of conventional continuum designs and more than twenty-seven times greater than typical magnet-tip robots, thereby establishing a new state of the art among 31 existing designs. Building upon this advance, we develop a Bio-Inspired Teleoperated Magnetic Navigation System (BITMNS) that integrates the model-guided MSR with a robotic magnetic manipulator and delivery device. In-vivo rabbit experiments demonstrate precise navigation through tortuous vascular pathways to the renal artery, while remote operation eliminates operator radiation exposure (238 mGy <inline-formula> <tex-math>$rightarrow ~0$ </tex-math></inline-formula>) and force analyses confirm robustness under loads up to 9.5 times the safety threshold. Collectively, these advances expand the navigable vascular workspace, eliminate operator radiation exposure, and establish a safety margin nearly one order of magnitude above clinical thresholds, thereby demonstrating a concrete pathway toward safer and more effective endovascular interventions.","PeriodicalId":73318,"journal":{"name":"IEEE transactions on medical robotics and bionics","volume":"8 2","pages":"654-666"},"PeriodicalIF":4.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148505159","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Design of a Variable-Stiffness Elbow Rehabilitation Robot and Its Control: A Variable Impedance Strategy Based on LSTM Elbow Torque Prediction","authors":"Ao Meng;Zhi Pang;Fengchen Liu;Baogui Qiu;Bingshan Hu;Hongliu Yu","doi":"10.1109/TMRB.2026.3680929","DOIUrl":"https://doi.org/10.1109/TMRB.2026.3680929","url":null,"abstract":"In the rehabilitation training of stroke patients, traditional rigid-driven robots struggle to effectively address spasticity, leading to suboptimal training and risks of secondary injury. This paper proposes a dual-motor driven variable stiffness elbow Joint rehabilitation robot based on a cam-spring-roller mechanism, enabling independent control of stiffness and position. We developed an LSTM-based elbow joint torque estimation model trained on a dataset generated using a musculoskeletal model incorporating Hill muscle model, achieving accurate and reliable torque prediction. A stiffness and impedance cooperative control framework is then established based on the estimated torque values: wherein the system provides higher assistive torque with lower mechanical stiffness when lower muscle strength is detected, and reduces assistive torque while increasing mechanical stiffness when higher muscle strength is identified. The framework also includes a spasticity detection model is proposed, incorporating metrics of antagonist muscle abnormal activation, normalized jerk score and velocity-activation quantification, aiming to enhance training safety and demonstrating the potential for improved patient engagement. Finally, experiments on four healthy subjects and a patient with spasticity demonstrated the feasibility and preliminary efficacy of the stiffness-variable impedance control framework and the spasticity detection model in a pilot study.","PeriodicalId":73318,"journal":{"name":"IEEE transactions on medical robotics and bionics","volume":"8 2","pages":"798-810"},"PeriodicalIF":4.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148507805","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Progress Toward Underwater Robotic Gait Assistance—A Review","authors":"Rishiraj Bose;Frank C. Sup","doi":"10.1109/TMRB.2026.3680958","DOIUrl":"https://doi.org/10.1109/TMRB.2026.3680958","url":null,"abstract":"This review examines the use of hydrotherapy for gait rehabilitation and analyzes the engineering considerations for underwater gait assistance. We summarize existing underwater gait systems and describe representative design choices in actuation, sensing, modeling, and control. We relate water-specific mechanics, including buoyancy, drag, and added mass, to metric selection in gait characterization and to requirements for estimation and feedback. The review compares modeling approaches for control, highlighting the tradeoff between fidelity and real-time use, and examines the practicality of actuation options under submersion conditions. Recurring gaps include the lack of direct underwater ground-reaction force measurements, limited phase estimation methods, and sparse validation data for controllers. We outline design considerations that follow from these findings and identify research needs in sensing, estimation, compliant actuation, and evaluation protocols. These findings link the effects of hydrotherapy to specific engineering choices that can be evaluated in rehabilitation settings.","PeriodicalId":73318,"journal":{"name":"IEEE transactions on medical robotics and bionics","volume":"8 2","pages":"612-626"},"PeriodicalIF":4.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148507938","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}