IEEE Transactions on Neural Systems and Rehabilitation Engineering最新文献

{"title":"Guidance Framework for Selecting Virtual Hand Illusion Paradigms to Enhance Motor Imagery via Sense of Ownership in Stroke Rehabilitation.","authors":"Hojun Jeong, Haemin Jung, Seyoung Shin, MinYoung Kim, Jonghyun Kim","doi":"10.1109/TNSRE.2026.3691422","DOIUrl":"https://doi.org/10.1109/TNSRE.2026.3691422","url":null,"abstract":"<p><p>Virtual hand illusion (VHI)-based motor imagery (MI) guidance systems are a promising approach for enhancing MI by reinforcing the sense of ownership (SoO), a key factor in effective neurorehabilitation. Although VHI-based guidance has shown potential, most prior studies have relied on fixed paradigms, limiting individual adaptability. This study investigates the mechanistic feasibility of a guidance framework for VHI paradigm selection based on patients' clinical characteristics and SoO responses to enhance MI-related neural activity in stroke patients. Twelve subacute stroke patients completed a two-visit protocol. During the initial visit, paradigm feasibility was assessed, and galvanic skin response (GSR) was recorded to support VHI selection when multiple paradigms were feasible. The final paradigm determined by the proposed framework was termed the guidance-selected VHI (GS-VHI). To validate the proposed method, SoO and MI enhancement were evaluated and compared under pure MI, action observation (AO), and GS-VHI conditions. In participants with multiple feasible VHI paradigms, the paradigm associated with stronger GSR-based SoO also elicited greater MI-related event-related desynchronization (ERD). GS-VHI outperformed conventional conditions in both SoO and MI enhancement, supported by EEG beta attenuation for SoO and by both channel- and source-level ERD analyses for MI. Additional relationship analyses further suggested that ownership-related indices were associated with MI enhancement. These findings highlight the potential of tailoring VHI paradigms using objective neurophysiological markers and support the mechanistic link between SoO and MI modulation. This study serves as a mechanistic feasibility study rather than a clinical efficacy trial.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147856290","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using EMG Biofeedback to Restore Closed-Loop Neural Control on a Powered Prosthetic Ankle.","authors":"Brendan Driscoll, Joshua R Tacca, Kasey Preisser, Ming Liu, Jonathan Stallrich, He Huang","doi":"10.1109/TNSRE.2026.3691653","DOIUrl":"https://doi.org/10.1109/TNSRE.2026.3691653","url":null,"abstract":"<p><p>This paper aims to present that biofeedback based on electromyography (EMG) can help transtibial amputees to improve their capability to manipulate powered prosthetic ankles using Direct Electromyography (dEMG) control. First, we constructed a novel haptic based EMG biofeedback system composed of a HD haptic vest and an encoder that mapped EMG magnitude of residual shank muscles to tactile vibration patterns. Then we recruited six transtibial amputees and six no-disabled participants to conduct a position matching task, in which they conducted dEMG control on a powered prosthetic ankle with/without the biofeedback system after a brief acclimation. Impact of the EMG biofeedback system was evaluated based on their accuracy in matching specific targets in the position matching task. Our results demonstrated that 1) the EMG biofeedback can improve participants' target matching accuracy; 2) the biofeedback led to more separable EMG signals among amputee participants when they tried to reach different targets; 3) for amputees, the improvement is more clearly observed when matching towards dorsiflexion targets; and 4) high performance variation among amputees is observed. Our work also demonstrated the feasibility of providing effective proprioceptive feedback based on muscle contraction level without involving any surgical procedure on lower limb amputees.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147856368","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Robust decomposition of surface EMG signals via lightweight deep learning-based adaptation.","authors":"Zeyu Zhou, Yang Yu, Yang Xu, Xinjun Sheng","doi":"10.1109/TNSRE.2026.3691346","DOIUrl":"https://doi.org/10.1109/TNSRE.2026.3691346","url":null,"abstract":"<p><p>Real-time surface electromyography decomposition has emerged as a promising way for neural interfacing. However, the decomposition performance faces dramatic degradation when multiple non-stationary factors coexist, including noise increases, new MU recruitments, and MU property variations. Here, we propose a deep learning-based (DL-based) adaptive decomposition method for moderate non-stationary scenarios, with an online adaptation strategy dynamically updating DL-based decomposition models. As prerequisite, the DL architecture is lightened through Tree-structured Parzen Estimator-based search to enable online adaptation. Additionally, multi-factor data augmentation was designed to enhance generalization capabilities. The proposed method outperforms blind source separation-based methods with noise increased by 5 dB, with F1-score of 0.715±0.227 vs. 0.388±0.342 and 0.633±0.047 vs. 0.202±0.052 for simulated and experimental data, respectively. Compared with DL-based methods, the proposed method is able to decode newly recruited MUs, with 11 and 6.93±3.04 MUs for simulated (total of 50 newly recruited MUs) and experimental data, respectively. Meanwhile, it performs better on initial MUs with property variations, with F1-score of 0.647±0.082 vs. 0.538±0.156 for experimental data. Regarding firing rates, the proposed method can identify a greater number of physiologically plausible MU spike trains. When deployed on an edge device, the proposed method met real-time decomposition (≤ 0.25 s) and online adaptation latency constraints. The outcomes demonstrate the effectiveness of lightweight DL-based adaptation for non-stationary sEMG decomposition, thus paving the way for MU-based neural interfacing.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147856326","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A 3D-Printing-Based Optogenetic Neural Stimulator Integrated with Three Neural Recording Channels.","authors":"Keonghwan Oh, Jihong Lim, Yehhyun Jo, Jiwan Woo, Yakdol Cho, Soon-Jae Kweon, Hyunjoo Jenny Lee, Sohmyung Ha","doi":"10.1109/TNSRE.2026.3690567","DOIUrl":"https://doi.org/10.1109/TNSRE.2026.3690567","url":null,"abstract":"<p><p>This paper presents an optical neural modulation device integrated with neural recording channels. While most optogenetic devices are fabricated using microfabrication techniques, this device is produced primarily using 3D printing, eliminating the need for complex microfabrication processes. The device largely comprises two layers: one for placing and connecting a light-emitting diode (LED) and another for integrating recording channels, both of which are fabricated using 3D printing based on two-photon polymerization (2PP) technology. The LED is placed on the 3D-printed substrate with conductive material positioned underneath and electrically connected with solder balls. The recording-channel layer has grooves to accommodate three recording channels made of platinum wires. The LED and recording layers are aligned and assembled using 3D-printed holes and trenches on each substrate layer. The device properties are characterized, including electrochemical impedance spectroscopy (EIS) of each recording channel. Finally, the device is implanted into the brain of a mouse, and neural signals are recorded while being stimulated optically with the same device.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147837316","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Adaptive Gait Assistance for Foot Drop Rehabilitation Based on Uncertainty Fusion of Contralateral Limb Information.","authors":"Kehan Xu, Jun Huo, Yize Zheng, Zixin Chi, Yu Cao, Zhaohui Yang, Jian Huang","doi":"10.1109/TNSRE.2026.3690647","DOIUrl":"https://doi.org/10.1109/TNSRE.2026.3690647","url":null,"abstract":"<p><p>Foot drop resulting from neurological injury severely compromises mobility and gait stability, yet existing assistive solutions often overlook physiological bilateral coordination and lack adaptability to individual gait variability. This study introduces a novel adaptive gait assistance framework for a soft exosuit that simultaneously aims to restore inter-limb coordination and generate personalized ankle joint assistance. First, we design a contralateral-guided adaptive phase synchronization engine that transforms healthy-limb information into an ideal phase reference for the impaired limb, enabling active correction of gait asymmetry. Second, to reconcile stability and personalization, we propose an uncertainty fusion-based adaptive gait assistance that estimates the real-time confidence of each model, arbitrates and fuses their predictions, and produces assistance trajectories that are both robust and individualized. After demonstrating high predictive accuracy and robustness in validation experiments with healthy subjects, a clinical evaluation involving five individuals with foot drop showed significant improvements in ankle kinematics, walking speed, and gait symmetry. Subjective feedback confirmed superior comfort, balance, and confidence over baseline and preset assistance. These preliminary results suggest that the proposed framework can improve gait kinematics, symmetry, and user-perceived walking performance, and support its potential as a personalized assistance strategy for foot-drop rehabilitation.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147837281","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel feedback-based compensation reduction with upper body reconstruction for upper-limb rehabilitation.","authors":"Yeji Hwang, Taegyun Kim, James Hyungsup Moon, Anna Lee, Hyo Jin Jeon, Jonghyun Kim, Min Ho Chun","doi":"10.1109/TNSRE.2026.3689896","DOIUrl":"https://doi.org/10.1109/TNSRE.2026.3689896","url":null,"abstract":"<p><p>Compensatory movements frequently occur during upper-limb rehabilitation for patients with stroke, potentially impeding effective motor recovery. Vision-based systems offer practical solutions for monitoring such compensations, but their application has often been limited by tracking inaccuracies and inadequate integration of trunk and arm movements. To address this limitation, we developed a real-time compensation detection and feedback system for upper-limb rehabilitation robot, which integrates upper-body reconstruction based on sensor-fusion with a machine learning classifier. The system estimates joint kinematics of both the trunk and arm to detect compensatory movement and delivers audio-visual feedback to reduce compensatory movement. We validated the system with 18 patients, who were divided into an experimental group and a control group, during robot-assisted planar reaching tasks. The classifier achieved reliable binary classification performance, with an F1-score 0.85 during classifier training and an F1-score 0.77 during real-time application with stroke. Importantly, the experimental group (n=9) significantly reduced compensatory movement occurrence from 58.9% at baseline to 38.2% post-training (p=0.016), showed improved trajectory mean distance from the theoretical path (p=0.0052), and exhibited decreased trunk movement magnitude (p<0.05). These results demonstrated effectiveness in reducing compensation and enhancing movement quality, highlighting its potential clinical utility for stroke rehabilitation.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147837303","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Validated Framework for Decoding Motor Unit Firings and Resulting Ankle Moments during Walking.","authors":"Antonio Gogeascoechea, Marco Carbonaro, Nathan Van Dieren, Utku S Yavuz, Massimo Sartori","doi":"10.1109/TNSRE.2026.3689740","DOIUrl":"https://doi.org/10.1109/TNSRE.2026.3689740","url":null,"abstract":"<p><p>Understanding how the central nervous system controls complex movements, such as walking, remains a fundamental challenge. Although motor units (MUs) are well-studied in isometric tasks, their role in generating joint moments during functional dynamic movements is unclear, partly due to non-stationary conditions. Filling this knowledge gap is essential for studying neural control of walking at the cellular level and for guiding neurorehabilitation and robotic interventions. We developed a validated framework for decoding high-density electromyography (HD-EMG) into individual MU spike trains during walking. We assessed both static decoding and an adaptive algorithm that continuously tracks time-varying action potentials, validating the results against fine-wire intra-muscular EMG (iEMG). We then examined neuromechanical delays (NMD) across walking speeds and established MU-driven neuromusculoskeletal models to determine the biomechanical consequences of the decoded MU firing patterns. Five healthy adults walked at multiple speeds while we recorded HD-EMG, iEMG, motion capture, and ground reaction forces. Both static and adaptive decompositions yielded comparable MU spike trains. Median rate of agreement was higher and false positives were lower for the adaptive approach, whereas false negatives were lower for the static approach. Although not statistically significant (p>0.05), these trends suggest the adaptive method may better handle nonstationarities. NMD decreased with speed, indicating coherent acceleration of neural-to-mechanical transmission. MU-driven models reproduced ankle joint moments with substantially lower normalized RMSE than conventional EMG-envelope models (p<0.05), with no difference between static and adaptive models. The close match between MU-derived and measured joint moments confirms that the decoded neural drive captures functional motor control rather than numerical artifacts. This study validates MU decomposition during walking and establishes a MU-driven model-based framework for investigating spinal motor control under dynamic, functionally relevant conditions. By bridging cellular-level neural activity with biomechanical outcomes, our approach opens new avenues for advancing neuro-rehabilitation, assistive technology, and our understanding of human locomotion.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147814357","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A 3-DoF wrist control based on natural arm movements outperforms current myoelectric prosthesis in VR.","authors":"Lucas Bardisbanian, Vincent Leconte, Emilie Doat, Remi Klotz, Aymar De Rugy","doi":"10.1109/TNSRE.2026.3689147","DOIUrl":"https://doi.org/10.1109/TNSRE.2026.3689147","url":null,"abstract":"<p><p>While mechatronics is progressing to over-come poor wrist capabilities of most current prosthetic devices, an efficient control system for a full 3 degrees-of-freedom (DoF) wrist is still lacking.We showed recently that novel controls based on Artificial Neural Network (ANN) trained on natural arm movements can predict multiple distal joints so well that participants with a transhumeral arm amputation could use them to reach objects as well as with a natural arm in virtual reality (VR). Here, we adapted this control to the case of transradial amputation, included important changes necessary for real-life applications, and compared it to currentmyoelectric control on two functional tasks (pick-and-place and clothespin relocation) performed in VR. When mechanical constraints of typical actual prostheses were simulated on participants without upper limb loss using a wrist brace (Exp1, n=20), success rates and movement times were only slightly degraded, but this was at the expense of large compensatory movements. When our 3-DoF wrist control was applied, good performances were maintained together with a dramatic reduction of those large compensatory movements. Participants with a transradial amputation (Exp2, n=8) had much lower performances with their prosthesis than with their intact arm, and benefited markedly from our wrist 3-DoF control both in terms of improved performances and reduced compensatory movements. These results demonstrate that the proposed movement-based 3-DoF wrist control outperforms current myoelectric prostheses in VR. This motivates further efforts needed toward the application to a real prosthesis.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147814034","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"NEuRT: A Transformer-Based Model for Explainable Neuronal Activity Analysis.","authors":"Georgii Raev, Daniil Baev, Evgenii Gerasimov, Viacheslav Chukanov, Ekaterina Pchitskaya","doi":"10.1109/TNSRE.2026.3689342","DOIUrl":"https://doi.org/10.1109/TNSRE.2026.3689342","url":null,"abstract":"<p><p>The study of neuronal activity is essential for understanding brain function and its alterations in neurode-generative diseases. Advances in in vivo imaging have enabled real-time observation of neuronal dynamics, but classical statistical methods struggle to capture the complex, time-dependent interactions within neuronal networks. Machine learning offers promising solutions for analyzing high-dimensional neuronal data, yet their application in neuroscience remains limited. Here, we introduce NEuRT, a Bidirectional Encoder Representations from Transformers (BERT)-based model adapted for neuronal activity analysis. NEuRT leverages self-attention mechanisms to interpret complex neuronal interactions, providing insights into patterns that traditional methods may overlook. Pre-trained on the recently introduced large annotated dataset MICrONS for signal reconstruction, NeuRT demonstrates strong generalization, effectively reconstructing activity from both visual cortex two-photon and hippocampal miniature fluorescence microscopy. Built on the BERT architecture, the NEuRT model can be efficiently fine-tuned for a wide range of downstream tasks. We showcase its application in classifying wild-type and transgenic Alzheimer's disease model mice, based on hippocampal activity, revealing group-specific features through attention map analysis. By reducing reliance on extensive labeled data, addressing a critical challenge in neuroscience, NEuRT bridges fundamental neuroscience and disease research, offering a robust framework for AI-driven and explainable neuronal activity analysis.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147814369","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"3D markerless tracking of speech movements with submillimeter accuracy.","authors":"Austin Lovell, James Liu, Arielle Borovsky, Raymond A Yeh, Kwang S Kim","doi":"10.1109/TNSRE.2026.3688903","DOIUrl":"10.1109/TNSRE.2026.3688903","url":null,"abstract":"<p><p>Speech movements, generated through precise spatial and temporal control of articulators, are inherently complex. Measuring these movements is challenging and often requires the use of multiple physical sensors positioned around the mouth and face to acquire precise movement measurements. Facial sensor placement can be difficult for certain populations to tolerate, particularly young children. Recent progress in machine learning-based markerless facial landmark tracking technology has demonstrated potential to provide lip tracking without the need for physical sensors, but whether such technology can provide submillimeter precision and accuracy in 3D remains unclear. Here, we developed a novel approach that integrates a facial landmark detector and CoTracker, a transformer-based neural network model that jointly tracks dense points across a video sequence. We further examined and validated this approach by assessing its tracking precision and accuracy. The findings revealed that our approach was more precise (≈ 0.15 mm in standard deviation) than a facial landmark detector alone (> 0.3 mm). In addition, its 3D tracking performance was comparable to electromagnetic articulography (≈ 0.3 mm RMSE against simultaneously recorded articulograph data). Importantly, the approach performed similarly well across adults and young children (i.e., 3- and 4-year-olds). Our novel framework leverages open-source pre-trained models, promoting accessibility and open science while using commercial-grade compute resources. It also serves as a proof of concept for improving the performance of a broad range of commonly used markerless tracking applications in neuroscience.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147814102","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}