IEEE Transactions on Neural Systems and Rehabilitation Engineering最新文献

{"title":"On the Benefit of FMG and EMG Sensor Fusion for Gesture Recognition Using Cross-Subject Validation","authors":"Maurice Rohr;Jad Haidamous;Niklas Schäfer;Stephan Schaumann;Bastian Latsch;Mario Kupnik;Christoph Hoog Antink","doi":"10.1109/TNSRE.2025.3543649","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3543649","url":null,"abstract":"Hand gestures are a natural form of human communication, making gesture recognition a sensible approach for intuitive human-computer interaction. Wearable sensors on the forearm can be used to detect the muscle contractions that generate these gestures, but classification approaches relying on a single measured modality lack accuracy and robustness. In this work, we analyze sensor fusion of force myography (FMG) and electromyography (EMG) for gesture recognition. We employ piezoelectric FMG sensors based on ferroelectrets and a commercial EMG system in a user study with 13 participants to measure 66 distinct hand movements with 10ms labelling precision. Three classification tasks, namely flexion and extension, single finger, and all finger movement classification, are performed using common handcrafted features as input to machine learning classifiers. Subsequently, the evaluation covers the effectiveness of the sensor fusion using correlation analysis, classification performance based on leave-one-subject-out-cross-validation and 5x2cv-t-tests, and its effects of involuntary movements on classification. We find that sensor fusion leads to significant improvement (42% higher average recognition accuracy) on all three tasks and that both sensor modalities contain complementary information. Furthermore, we confirm this finding using reduced FMG and EMG sensor sets. This study reinforces the results of prior research about the effectiveness of sensor fusion by performing meticulous statistical analyses, thereby paving the way for multi-sensor gesture recognition in assistance systems.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"935-944"},"PeriodicalIF":4.8,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10892285","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143535527","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Throw and Catch: Analyzing the Synchronized Movements of Eyes and Joints in Children","authors":"Ziyu He;Hong Fu;Ruimin Li;Zhen Liang;Chetwyn C. H. Chan;Yanwen Xu;Yang Zheng","doi":"10.1109/TNSRE.2025.3543730","DOIUrl":"10.1109/TNSRE.2025.3543730","url":null,"abstract":"Throw and catch are fundamental motor skills that are closely related to eye-hand coordination, reaction speed, and spatial awareness in children. Current research on throw and catch mainly focuses on the impact of attentional focus, anticipatory knowledge, and training on visuomotor control. Little work has been done on the synchronized movements of eyes and joints during the throw and catch. To understand how these synchronized movements contribute to the success rate of throwing and catching, we proposed a video-based framework named Synchronized Eye and Joint Analysis (SEJA). This framework locates, extracts, and analyzes the essential eye and joint movements from untrimmed first-person and third-person view videos. Using the proposed framework, throw and catch events in long untrimmed videos were successfully identified, and whether each catch was successful was accurately assessed. Additionally, detailed metrics related to predictive gaze behaviors and predictive hand movements for each catch event were obtained. On a dataset consisting of videos from 56 children aged 7 to 10, the proposed framework delivered an average precision (AP) ranging from 0.5 to 0.95 at 0.881 for task localization and achieved an accuracy of 0.985 in predicting whether a catch was successful. Our research indicated that children with higher catch success rates showed shorter delays in predicting the ball’s trajectory, smaller amplitudes of body movement, and more pronounced predictive saccades (rapid eye movements to anticipate the ball’s position). These findings are crucial for comprehending and improving the development of motor skills in children.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"994-1003"},"PeriodicalIF":4.8,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10892245","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143556759","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impact of Visual Clutter in VR on Visuomotor Integration in Autistic Individuals","authors":"Minxin Cheng;Leanne Chukoskie","doi":"10.1109/TNSRE.2025.3543131","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3543131","url":null,"abstract":"Autistic individuals often exhibit superior local visual sensitivity but may struggle with global visual processing, affecting their visuomotor integration (VMI). Goal-directed overhand throwing is common in both the physical environment (PE) and virtual reality (VR) games, demanding spatial and temporal accuracy to perceive position and motion, and precise VMI. Understanding VMI in autistic individuals and exploring supportive designs in VR are crucial for rehabilitation and improving accessibility. We assessed static visuospatial accuracy and VMI with autistic (<inline-formula> <tex-math>${n} = 16$ </tex-math></inline-formula>) and non-autistic (<inline-formula> <tex-math>${n} = 16$ </tex-math></inline-formula>) adults using spatial estimation and overhand throwing tasks with eye and hand tracking, comparing VR to PE. In VR, all participants exhibited reduced visual accuracy, increased visual scanning, and shortened quiet eye duration and eye following duration after the ball release, which led to decreased throwing performance. However, simplifying visual information in VR throwing improved these measures, and resulted in autistic individuals outperforming non-autistic peers.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"829-840"},"PeriodicalIF":4.8,"publicationDate":"2025-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10891367","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143480803","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Statistical Multiscore Functional Atlas Creation for Image-Guided Deep Brain Stimulation","authors":"Xiongbiao Luo;Zhuo Zeng;Song Zheng;Jianhui Chen;Pierre Jannin","doi":"10.1109/TNSRE.2025.3542395","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3542395","url":null,"abstract":"Deep brain stimulation is increasingly performed for patients who suffer from drug-resistant movement disorders. It still remains challenging to determine the optimal electrode contact location to obtain the optimal surgical outcome and simultaneously minimize adverse effects. This paper proposes to construct a new statistical functional atlas to guide electrode contact targeting during deep brain stimulation. The construction of the atlas consists of four main steps: 1) multimodal image segmentation and registration, 2) activation volume modeling, 3) computing and combining multiple functional scores, and 4) generation of multiscore functional atlas. Based on these steps, the statistical functional atlas is created by integrating anatomical information analysis with multiple clinical scores that postoperatively characterize stimulation efficacy (e.g., motor symptom) and adverse effect. We evaluated the created atlas on 40 subthalamic nucleus stimulated parkinsonian patient datasets. The experimental results show that the reproducibility of the created statistical functional atlas was more than 75% in the cross validation. In addition, the motor, neuropsychological, and health scores can be reproduced up to 77%, 82%, and 78%. Compared to the actually implanted electrode position, the atlas predicted and the manually planned electrode position errors were 2.89 mm and 2.38 mm, respectively. The constructed multiscore atlas provides an automatic and accurate electrode targeting strategy that potentially outperforms manually planned approaches.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"818-828"},"PeriodicalIF":4.8,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10891035","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143465667","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Estimating Gait Speed in the Real World With a Head-Worn Inertial Sensor","authors":"Paolo Tasca;Francesca Salis;Samanta Rosati;Gabriella Balestra;Claudia Mazzà;Andrea Cereatti","doi":"10.1109/TNSRE.2025.3542568","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3542568","url":null,"abstract":"Head-worn inertial sensors represent a valuable option to characterize gait in real-world conditions, thanks to the integration with glasses and hearing aids. Few methods based on head-worn sensors allow for stride-by-stride gait speed estimation, but none has been developed with data collected in real-world settings. This study aimed at validating a two-steps machine learning method to estimate initial contacts and stride-by-stride speed in real-world gait using a single inertial sensor attached to the temporal region. A convolutional network is used to detect strides. Then, stride-by-stride gait speed is inferred from the detected cycles by a gaussian process regression model. A multi-sensor wearable system was used to label over 100,000 strides recorded from 15 healthy young adults during supervised acquisitions and real-world unsupervised walking. The stride detector achieved high detection rate (F1-score > 92%) and accuracy (mean absolute error < 40 ms). Very strong correlation between target and predicted speed (Spearman coefficient > 0.86) and low mean absolute error (< 0.085 m/s) were observed. The method proved valid for the quantitative evaluation of stride-by-stride gait speed in real-world conditions. These findings lay the technical and analytical groundwork for future clinical validation and application of gait analysis frameworks that integrate inertial sensors with head-worn devices.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"858-867"},"PeriodicalIF":4.8,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10891015","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143496593","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Adaptive Personalized Ankle-Foot-Orthosis: 3D Printing and On-Site Carbon Fiber Reinforcement for Tailored Stiffness","authors":"Janna Krummenacker;Ulrich Blass;Nicole Motsch-Eichmann;Joachim Hausmann","doi":"10.1109/TNSRE.2025.3542116","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3542116","url":null,"abstract":"The current manufacturing process of ankle-foot orthoses (AFOs) allows for little post-processing and subsequent patient customization. This study aims to answer the question of how to provide faster, more efficient, and more individualized patient care. The first step is to develop a design for an AFO that can be 3D printed for each patient and then reinforced with thermoplastic fiber tape. This design is then analyzed using a finite element model (FEM). With this model, the mechanical performance of the reinforced AFO is compared to that of the pure 3D printed AFO. At the same time, the entire AFO manufacturing process chain was redesigned and adapted to the new design. The experimental and computational results are in good agreement and show a significant improvement in stiffness due to the tape reinforcement. The result of this study is the development of a lightweight AFO design that can be easily and efficiently fitted to the patient in the field.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"841-846"},"PeriodicalIF":4.8,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10887337","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143496466","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhanced Functionality Using a Powered Upper Extremity Exoskeleton in Patients With Brachial Plexus Injuries","authors":"Sandesh G. Bhat;Emily J. Miller;Paul Kane;Kevin W. Hollander;Claudio Vignola;Alexander Y. Shin;Thomas G. Sugar;Kenton R. Kaufman","doi":"10.1109/TNSRE.2025.3538175","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3538175","url":null,"abstract":"Traumatic brachial plexus injury (BPI) results in significant disability, often hindering functionality in the patient’s daily life. Post- surgery, muscle strength recovery can take up to two years, with 40% of patients requiring even longer. A powered elbow orthosis can enhance functionality during activities of daily living (ADLs). This study tested a novel powered myoelectric elbow orthosis (PMEO) during ADLs. Subjects with BPI were fitted with the PMEO and divided into two groups: more impaired (Manual Muscle Test (MMT) < 3, N = 5) and less impaired (MMT≤ 3, N = 4). They performed four ADLs involving full elbow motion, including an activity requiring the subjects to lift a basket with weights. Upper extremity kinematics, electromyographic activity, weight lifted, and subject feedback on the device’s form and fit were collected and analyzed. Results showed that the PMEO significantly improved elbow range of motion in the more impaired group (14 ± 23 degrees, p = 0.019) without any additional compensatory motions in the shoulder or trunk. More impaired subjects lifted an average of 1.1 ± 0.6 kg with the PMEO, whereas they could not do so without it (p = 0.011). Subjects appreciated the PMEO’s weight, fit, and form. All could don and doff the device with minimal assistance. These findings demonstrate that the PMEO is a viable option to enhance ADL function for patients with BPI.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"780-786"},"PeriodicalIF":4.8,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10887042","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143465665","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"High-Density EMG Grip Force Estimation During Muscle Fatigue via Domain Adaptation","authors":"Huiming Pan;Dongxuan Li;Chen Chen;Peter B. Shull","doi":"10.1109/TNSRE.2025.3541227","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3541227","url":null,"abstract":"Myoelectric interfaces hold promise for enabling intuitive and natural control of prostheses and exoskeletons. Muscle fatigue, whether due to prolonged use or heavy weight loads, can alter the distribution of electromyographic (EMG) signals, leading to inconsistencies compared to non-fatigued conditions. This presents significant challenges for accurately decoding user intentions. We thus propose a novel estimation method based on domain adaptation to improve grip force estimation accuracy during muscle fatigue. Specifically, the proposed method integrates an adversarial training strategy and an end-to-end deep learning model to align EMG feature distributions across non-fatigue and fatigue states. A baseline model, whose structure was identical to the force estimation network of the proposed method, was used for comparison. Eight subjects performed non-fatigue and fatigue gripping tasks, and grip force estimations were compared with dynamometer gold standard measurements. Results demonstrate that root mean square error (RMSE) of the proposed model was 51.9% lower than that of the baseline model during muscle fatigue. The proposed method leverages labeled data in the non-fatigue domain and employs adversarial objectives to ensure the learned features are applicable to both domains, which eliminates the need to pause to collect force label data in the fatigue domain, expediting and simplifying the calibration process. This study has the potential to improve the ability of myoelectric interfaces during muscle fatigue to enable users to intuitively retrieve and grip objects over extended periods, ultimately improving independence and quality of life.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"925-934"},"PeriodicalIF":4.8,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10884597","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143535472","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Signal Processing and Feature Extraction in Markerless Telerehabilitation","authors":"Silvana G. Dellepiane;Federica Ferraro;Camilla Baffigo;Marina Simonini","doi":"10.1109/TNSRE.2025.3541153","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3541153","url":null,"abstract":"Telerehabilitation solutions are a concrete answer to many needs in the healthcare framework since they enable remote support for patients and foster continuity of care. This paper explores telerehabilitation using the ReMoVES system, a markerless approach that facilitates remote exercise guidance. Focusing on the sit-to-stand (STS) task, which is crucial for daily activities, this study employs the Microsoft Kinect sensor for human movement monitoring. Emphasizing preprocessing and analysis, the research extracts reliable parameters, enabling remote observation and evaluation of patient performance. This study highlights the importance of noise reduction and automatic segmentation for feature extraction, which are essential for assessing task execution and identifying compensatory movements. By utilizing a diverse healthy subject group, a reference model is established, providing optimal features for accurate exercise execution. Statistical analyses involving both healthy subjects and patients revealed key features for remote exercise observation. Automatic feature extraction related to poses and body movements, together with homogeneity within control group sessions, forms the basis for a quantitative parametric model. This model describes and compares accurate exercise execution, offering a method to remotely evaluate and adapt individual rehabilitation plans on the basis of robust and reliable parameters.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"911-924"},"PeriodicalIF":4.8,"publicationDate":"2025-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10879802","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143535471","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Continuous Estimation of Swallowing Motion With EMG and MMG Signals","authors":"Zhenhui Guo;Ziyang Wang;Yue Wang;Weiguang Huo;Jianda Han","doi":"10.1109/TNSRE.2025.3540842","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3540842","url":null,"abstract":"Oropharyngeal dysphagia (OD) is a symptom of swallowing dysfunction that is associated with aspiration, severe respiratory complications, and even death. OD is a highly prevalent condition in populations including the elderly and patients with neurological diseases (e.g., stroke and Parkinson’s disease (PD)). Assessment of swallow function is crucial for managing OD, yet depends on devices for long-term monitoring during daily life and relevant methods for accurately assessing swallow function. The videofluoroscopic swallowing study (VFSS) is usually considered a gold standard method. However, it has several limitations, such as radiation exposure, the need for technical experts, high cost, and clinical use only. This study investigates the performances of electromyography (EMG) and mechanomyography (MMG) signals, which can be easily measured using wearable sensors, to continuously estimate swallowing movement. Meanwhile, three methods, i.e., Gaussian process regression (GPR), LSTM, and random forest (RF), are used for swallowing motion estimation based on EMG/MMG signals measured from six healthy subjects and a patient with PD, respectively. Moreover, a depth camera-based approach is proposed to provide the reference laryngeal displacement (i.e., the swallowing movement). The experimental results show that EMG models with three machine learning methods can accurately estimate swallowing movement. For the healthy subjects, the mean correlation coefficient (CC) is about 0.90 and the normalized root mean square error (NRMSE) is less than 0.15. For the PD patient, the CC is 0.804 and the NRMSE is 0.205 when using RF. The performance of the MMG model is comparable to that of EMG: CC/NRMSE of the LSTM model is 0.844/0.150 (healthy subjects); CC/NRMSE of RF model is 0.727/0.204 (PD patient). To the best of our knowledge, this is the first study proving that both EMG and MMG are two effective means for an accurate continuous estimation of swallowing motion, enabling the possibility of a safe and convenient evaluation and management of OD.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"787-797"},"PeriodicalIF":4.8,"publicationDate":"2025-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10879466","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143465664","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}