IEEE Transactions on Neural Systems and Rehabilitation Engineering最新文献

{"title":"Individualized Target Selection of Closed-loop Electrical Stimulation for the Treatment of Spontaneous Temporal Lobe Epilepsy.","authors":"Yufang Yang, Haoqi Ni, Yuting Sun, Yanjie Xing, Chang Wang, Jianmin Zhang, Junming Zhu, Kedi Xu","doi":"10.1109/TNSRE.2025.3614867","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3614867","url":null,"abstract":"<p><p>Closed-loop neuromodulation holds significant promise for treating refractory epilepsy, but the lack of specificity and individualization considerably limits its clinical efficacy. Given the inherent complexity of epilepsy, which involves multiple brain regions and significant interindividual variability, a network-guided, personalized approach is essential. This study aims to develop precise, individualized neuromodulation strategies by leveraging unique brain network characteristics. Using a closed-loop system in chronic temporal lobe epilepsy (cTLE) rats, continuous neural signals were analyzed to identify optimal stimulation targets via the Granger causality (GC) method. Results showed that brain network connectivity remained stable in the short term but changed significantly over time. GC-guided stimulation effectively reduced seizure duration, enhancing θ and α frequency band activity while suppressing γ activity. Additionally, targeted stimulation briefly inhibited interictal spikes and suppressed high-frequency oscillations during seizures. These findings highlight the potential for personalized neuromodulation to improve epilepsy treatment outcomes and deepen understanding of its underlying mechanisms.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2025-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145174907","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 Global-Local Dynamic Directed Graph Neural Network for Parkinson's Disease Detection.","authors":"Xiaotian Wang, Guanhai Zhou, Zhifu Zhao, Xiaoyi Zhang, Fu Li, Fei Qi","doi":"10.1109/TNSRE.2025.3614430","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3614430","url":null,"abstract":"<p><p>Graph Neural Networks (GNNs) for gait analysis utilizing Vertical Ground Reaction Force (VGRF) signals have demonstrated significant potential in Parkinson's Disease (PD) diagnosis and rehabilitation fields. However, existing GNN-based methods in this area normally model the VGRF signals as static topological structures, and ignore the dynamic variations of the VGRF graph structures during walking. To address this issue, a Global-Local Dynamic Directed Graph Neural Network (GLD²-GNN) is proposed to represent the dynamic spatio-temporal features of VGRF signals. The core component of the proposed model is the DyDGNN block, which is composed of a Dynamic Graph Learning (DGL) unit, a Dynamic Directed Graph Network (DyDGN) unit, and a Temporal Convolutional Network (TCN) unit. First, the DGL unit is proposed to learn dynamic topological relationships of VGRF signals. Based on learned graph structures, the DyDGN unit is constructed to extract the spatial patterns and capture topological dynamic features from VGRF signals. Subsequently, local temporal patterns of VGRF signals are extracted by the TCN unit. The proposed method is evaluated through k-fold cross-validation and cross-dataset validation on three datasets Ga, Ju and Si. Compared with existing methods, such as RFdGAD, Transformer, and AST-DGNN, GLD²-GNN demonstrates superior performance in the validation experiments. Notably, our method achieves an average improvement of 4.45% in accuracy, 2.93% in F1 score, and 2.88% in geometric mean across cross-dataset validation. Extensive experiments have demonstrated that GLD²-GNN exhibits both the representational ability for complex gait patterns and the generalization ability across various datasets by capturing dynamics of VGRF topological structures and spatio-temporal features from VGRF signals. For future work, we plan to combine our method with multi-modal methods and integrate our framework into a complete gait analysis system.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145148886","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":"Dynamic Stability Analysis of Progressive Supranuclear Palsy affected gait using Lyapunov Floquet Theory.","authors":"Sandesh G Bhat, Farwa Ali, Cecilia A Hogen, Asghar Rezaei, Keith A Josephs, Jennifer L Whitwell, Kenton R Kaufman","doi":"10.1109/TNSRE.2025.3614555","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3614555","url":null,"abstract":"<p><p>Progressive supranuclear palsy (PSP) is a neurodegenerative disease with severe gait and balance deficits. There are no effective ways to assess dynamic balance during walking in PSP. The Lyapunov Floquet (LF) theory has been utilized to study dynamic balance in healthy and pathologic gait but has not been applied to PSP affected gait. In the current study, the medio-lateral motion of the center of mass during gait for 40 patients with PSP (PSP group) and 33 healthy older adults (Control group) were studied. Metrics from LF theory, such as the maximal Floquet multiplier (FM), maximal long-term Lyapunov Exponent (LE_L), and maximal short-term Lyapunov Exponent (LE_S) were used to study walking stability. Although all the gait dynamics for all the participants were stable and non-chaotic, the PSP group was observed to be closer to an unstable system and more susceptible to perturbations (|FM| closer to 1 and LE_L closer to 0) than the Control group (p < 0.001). The control group's stability deteriorated, and the gait system became more susceptible to perturbations with age. Such a trend was not observed in the PSP group. The risk of falls increased with increase in cadence in the PSP group (p < 0.001). These findings demonstrate the potential of LF theory measures to evaluate dynamic stability in patients with PSP and the need for future research using quantitative measures.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145148910","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":"EEG Microstate Imbalance in Anti-LGI1 Encephalitis: A Correlation with Inflammation and Cognitive Dysfunction.","authors":"Xue Yang, Xiaotang Wang, Si Chen, Xiaxin Yang, Xiuhe Zhao","doi":"10.1109/TNSRE.2025.3614263","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3614263","url":null,"abstract":"<p><p>This study investigates changes in electroencephalogram (EEG) microstates in patients with anti-leucine-rich glioma-inactivated protein 1 (anti-LGI1) encephalitis and explores their correlation with inflammatory markers and cognitive function. Thirty patients with LGI1 encephalitis were compared to a control group of thirty healthy individuals, analyzing demographics, blood/cerebrospinal fluid (CSF) tests, and cognitive assessments. Each participant underwent at least 10 minutes of resting-state EEG recording. Microstate analysis and functional connectivity were performed using MATLAB's EEGLAB toolbox, while standardized low-resolution brain electromagnetic tomography (sLORETA) was employed for microstate source reconstruction. Patients with anti-LGI1 encephalitis exhibited prolonged durations of microstates A, B, and C, reduced occurrence of microstate D, and increased transitions from C to A, with no significant changes in coverage. Microstate A demonstrated decreased activity in the anterior cingulate and lingual gyri. Functional connectivity analysis revealed enhanced slow-wave connectivity and diminished fast-wave connectivity in the frontal-parietal lobes. Correlation analyses showed that microstate A positively correlated with inflammatory indices such as the neutrophil-to-lymphocyte ratio (NLR), systemic inflammation index (SII), and systemic inflammatory response index (SIRI). In contrast, microstate D correlated with the derived neutrophil-to-lymphocyte ratio (dNLR). Microstates B and C inversely correlated with NLR and SII. Furthermore, the duration and coverage of microstate A, along with the transition probability from A to D, inversely correlated with Mini-Mental State Examination (MMSE) scores. Similarly, the AD/BC ratio in occurrence, duration, and coverage also negatively correlated with MMSE scores. These findings revealed alterations in EEG microstates in patients with LGI1 encephalitis, highlighting an imbalance in the AD/BC ratio associated with inflammatory processes and cognitive impairments.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145148896","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":"Versatile Neural Activation Predictor with Axon Structure Tailoring Capability Enabling Personalized Neuromodulation Computation.","authors":"Hongda Li, Shunjing Wang, Xuesong Luo, Changqing Jiang, Boyang Zhang","doi":"10.1109/TNSRE.2025.3614215","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3614215","url":null,"abstract":"<p><p>Neuromodulation therapies are evolving to be more and more intelligent and personalized, driving the need for more precise and efficient stimulation strategies. Biophysically detailed computational models integrated with anatomically accurate neural structures could offer critical insights into neural activation patterns under various stimulation conditions, which are essential to optimize the treatment. However, solving these models containing a large number of nerve fibers is computationally intensive, especially when the neural targets comprise heterogenous axons, e.g., with varying geometries. Also, current methods lack generalizability across various neuromodulation scenarios, limiting the scalability and clinical utility of such models. In this study, we present a convolutional neural network (CNN)-based framework as a universal, rapid, and accurate alternative to conventional case-by-case brutal force computation methods. Our approach achieves a mean absolute error (MAE) of 6.91 × 10^-3 mV and over 95% prediction accuracy under diverse extracellular stimulation scenarios, facilitating personalized simulations and tailored neuromodulation treatments.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145148876","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":"Muscle Synergy-Based Iterative Learning Control for Upper Limb Functional Electrical Stimulation in Stroke Rehabilitation.","authors":"Yanhong Liu, Yaowei Li, Zan Zhang, Benyan Huo, Long Cheng, Anqin Dong, Gen Li","doi":"10.1109/TNSRE.2025.3613998","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3613998","url":null,"abstract":"<p><p>Functional Electrical Stimulation (FES) is widely used in the postoperative rehabilitation of stroke patients. Multi-channel FES enables alternating stimulation of multiple muscle groups, effectively delaying muscle fatigue and facilitating precise control of complex upper limb movements. However, high-dimensional control of multiple muscles introduces additional challenges, particularly in coordinating antagonistic muscles and achieving efficient control. This study proposes a novel FES control framework that integrates muscle synergy theory, Long Short-Term Memory (LSTM) networks, and Iterative Learning Control (ILC). In this framework, the LSTM network predicts synergy activation coefficients from joint kinematics (angle and angular velocity), while the ILC algorithm iteratively updates electrical stimulation intensities based on the tracking error from previous iterations. This combination reduces the dimensionality of muscle control and improves the balance of muscle group activation, aligning better with natural motor control strategies. Experiments conducted on eight healthy subjects demonstrated that the proposed synergy-based ILC method significantly reduced joint angle tracking errors (measured by RMSE) over 10 stimulation iterations, compared to reference trajectories derived from voluntary motion. Specifically, in the combined elbow-wrist drinking task, the wrist RMSE decreased from 13.10° to 4.19°, and the elbow RMSE decreased from 45.07° to 5.53°. The coefficient of determination (R²), reflecting the goodness of fit between predicted and reference trajectories, exceeded 0.96, indicating high tracking accuracy and stability. Preliminary experiments on three stroke patients further support the adaptability and clinical potential of the proposed method.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145137354","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":"Modelling and analysis of myoelectric networks based on lower limb motor synergies.","authors":"Lingling Chen, Yanglong Wang, Xulong Lu, Junjie Geng, Tengyang Feng","doi":"10.1109/TNSRE.2025.3613408","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3613408","url":null,"abstract":"<p><p>The synergy between muscles is a prerequisite for the human body to complete various movements, and it undergoes dynamic changes during the movement process, which cannot be characterized using traditional muscle synergy extraction methods. Therefore, it is necessary to establish an analytical framework that covers the entire dynamic process to decode inter-muscular interaction information completely. By extracting the time-domain characteristic values of EMG, a complex network model is established to analyze the EMG dynamics of lower limb from the nodes and edges of network. On the one hand, the nodes are community-detected with the goal of maximum modularity, and the EMG network is hard-divided into several discrete communities. On the other hand, the edges are matrix-decomposed to obtain different subgraphs over time. The surface EMG of 18 subjects were collected during passive and active training with rehabilitation robot. The experimental results show that the muscle synergy is most substantial during the flexion and extension phases of lower limb movement, with muscle groups mainly composed of the rectus femoris, lateral thigh muscles, and calf gastrocnemius. From the results of subgraph decomposition, it can be concluded that the synergistic effects between different regions of the legs are similar, and the synergistic ability between thigh muscles is significantly higher than that of calf muscles. The dynamic network modeling framework provides a new idea for muscle synergy analysis, which can not only analyze the community clustering of muscle groups from a macro perspective, but also quantify the degree of synergy between muscle regions.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2025-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145130813","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":"Continuous Reaching and Grasping with a BCI Controlled Robotic Arm in Healthy and Stroke-Affected Individuals.","authors":"Dylan Forenzo, Yisha Zhang, George F Wittenberg, Bin He","doi":"10.1109/TNSRE.2025.3611821","DOIUrl":"10.1109/TNSRE.2025.3611821","url":null,"abstract":"<p><p>Recent advancements in signal processing techniques have enabled non-invasive Brain-Computer Interfaces (BCIs) to control assistive devices, like robotic arms, directly with users' EEG signals. However, the applications of these systems are currently limited by the low signal-to-noise ratio and spatial resolution of EEG from which brain intention is decoded. In this study, we propose a motor-imagery (MI) paradigm, inspired by the mechanisms of a computer mouse, that adds an additional \"click\" signal to an established 2D movement BCI paradigm. The additional output signal increases the degrees of freedom of the BCI system and may enable more complex tasks. We evaluated this paradigm using deep learning (DL) based signal processing on both healthy subjects and stroke-survivors in online BCI tasks derived from two potential applications: clicking on virtual targets and moving physical objects with a robotic arm in a continuous reach-and-grasp task. The results show that subjects were able to control both movement and clicking simultaneously to grab, move, and place up to an average of 7 cups in a 5-minute run using the robotic arm. The proposed paradigm provides an additional degree of freedom to EEG BCIs, and improves upon existing systems by enabling continuous control of reach-and-grasp tasks instead of selecting from a discrete list of predetermined actions. The tasks studied in these experiments show BCIs may be used to control computer cursors or robotic arms for complex real-world or clinical applications in the near future, potentially improving the lives of both healthy individuals and motor-impaired patients.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145086015","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":"Quantitative Assessment and Diagnosis of Muscle Function in Sarcopenia Based on EIT-derived Parameters.","authors":"Yanan Diao, Guilan Chen, Junwen Peng, Nan Lou, Bo Sun, Jiafeng Yao, Guanglin Li, Guoru Zhao","doi":"10.1109/TNSRE.2025.3611827","DOIUrl":"https://doi.org/10.1109/TNSRE.2025.3611827","url":null,"abstract":"<p><p>The quantitative evaluation and diagnosis of muscle function in patients with sarcopenia are crucial to mitigate functional decline and the health burden in aging populations. This study proposed a method for the classification of sarcopenia and the evaluation of muscle function scores based on EIT technology. We recruited 31 participants, including individuals with sarcopenia (n = 11), healthy elderly (n = 10), and healthy young adults (n = 10), obtained muscle clinical fitness assessment scores and EIT-derived parameters, conducted intergroup comparisons of EIT parameters and clinical scores, and constructed a machine learning classification model for sarcopenia. EIT parameters conductivity (σ) were significantly different among the three groups (p < 0.05). Clinical muscle function scores showed a strong positive correlation with the σ (r = 0.73, R² = 0.54, p < 0.001), while negatively correlated with impedance (Z) (r = -0.55, R² = 0.27, p < 0.05). In addition, σ was positively correlated with hand grip strength (HGS) (r = 0.52, R² =0.20, p=0.30), and maximum voluntary muscle contraction (MVC) (r=0.73, R² = 0.49, p<0.001), and negatively correlated with age (r = -0.76, R² = 0.56, p<0.001) and SARC-F scale scores (r = -0.73, R² =0.57, p<0.001). Finally, the KNN-based sarcopenia classification model demonstrated strong performance in classification tasks, as evidenced by an accuracy of 0.89 and an AUC of 0.94. This study demonstrates that the EIT is a portable, wearable, and long-term monitoring tool for assessing and classifying muscle function in sarcopenia. With further clinical validation, it is expected to be used for early screening and rehabilitation monitoring of sarcopenia.</p>","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"PP ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145085995","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":"Training Movement Velocity Significantly Affects the Performance of Myoelectric Control","authors":"Troy N. Tully;Amelia E. Nelson;Jacob A. George","doi":"10.1109/TNSRE.2025.3610352","DOIUrl":"10.1109/TNSRE.2025.3610352","url":null,"abstract":"Our native hands are uniquely capable of operating across a wide range of speeds and forces. In contrast, most commercial myoelectric prostheses typically provide limited speed and force output. One approach to endow myoelectric prostheses with variable speed and/or force output is to use continuous kinematic positions of the prosthesis based on electromyography (EMG). Within the field of machine learning, it is well established that homogeneous training data can lead to bias that negatively impacts the run-time performance of the algorithm. Yet, most continuous decoders are trained on a homogeneous dataset involving only a single kinematic speed. To this end, we systematically investigated how different training speeds influence myoelectric control with two common continuous decoders on multiple performance metrics. We compared a Kalman filter (KF) and Convolutional Long Short-Term Memory (C-LSTM) neural network trained on slow, medium, fast, and mixed-speed datasets, evaluating their performance in offline analyses and in two real-time online tasks with the user actively in the loop. We found that training speed significantly affected algorithm performance, but effects were often algorithm dependent. Linear algorithms, like the KF, are likely to exhibit lower unintended movement errors and smoother control when trained on slow-speed data but will also struggle to generalize to higher movement speeds. In contrast, nonlinear algorithms like the C-LSTM can likely provide greater adaptability, with mixed-speed training leading to improved accuracy and task success rates across conditions. Although an often-overlooked implicit parameter, these findings explicitly demonstrate that a lack of diverse training speeds in existing myoelectric control training paradigms leads to worse decoder performance. By incorporating a range of movement speeds into training protocols or decoder design, myoelectric continuous decoders could achieve more dexterous and robust control, potentially improving prosthetic usability and retention.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"3784-3792"},"PeriodicalIF":5.2,"publicationDate":"2025-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11165468","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145075184","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}