IEEE Transactions on Neural Systems and Rehabilitation Engineering最新文献

{"title":"Enhancing Neural Representations of Motor Imagery Through Action-Specific Brain Connectivity Patterns","authors":"Guangying Wang;Lin Jiang;Xipeng Song;Yulin Zhang;Dezhong Yao;Jing Lu;Peng Xu;Fali Li;Yi Liang","doi":"10.1109/TNSRE.2025.3605612","DOIUrl":"10.1109/TNSRE.2025.3605612","url":null,"abstract":"Motor imagery (MI) is a cognitive process that allows individuals to mentally simulate movements without physical executio n. However, the exploration of functional connectivity (FC) and lateralization mechanisms under different MI actions remains insufficiently understood. In this work, the common orthogonal basis extraction (COBE) algorithm was employed to isolate action-specific components by removing shared background components from the raw FC of the MI process. We demonstrate that action-specific FC effectively captures the hemispheric statistical differences between left- and right-hand MI, outperforming traditional FC and temporal variability measures. And through a comprehensive analysis of network properties at three distinct levels, encompassing the whole-brain network properties, hemispherical properties, and individual nodal strength, complex lateralization patterns associated with diverse types of MI processes were successfully discerned. Furthermore, lateralization indices were further calculated to quantitatively reveal the degree of brain lateralization. Notably, the lateralization performance (LP) derived from action-specific FC exhibited a significant predictive capacity for MI performance, thereby suggesting its potential to evaluate individual MI capability. Collectively, these findings validate the action-specific FC patterns in characterizing neural mechanisms of MI processes and indicate that the LP could potentially be a useful tool to predict the MI performance of MI-based brain-computer inference (BCI), thereby contributing to the formulation of personalized therapeutic strategies for clinical rehabilitation from a new perspective.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"3555-3564"},"PeriodicalIF":5.2,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11150519","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144992138","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":"Predictive Force Control of Fingertip Induced by Functional Electrical Stimulation Based on a Hill-Type Muscle Model","authors":"Junyun Fu;Chengyu Lin;Jinxin Sun;Kong Hoi Cheng;Yuquan Leng;Chenglong Fu","doi":"10.1109/TNSRE.2025.3605816","DOIUrl":"10.1109/TNSRE.2025.3605816","url":null,"abstract":"Functional electrical stimulation (FES) is an effective technique for restoring or enhancing hand motor function in patients with neurological impairments, such as those recovering from stroke or spinal cord injuries. Although many studies have used phenomenological models to investigate the control of FES, few studies have simultaneously employed both methods to study finger output force. This study aims to accurately predict finger output force using the Hill model and a multi-joint finger model under different current conditions. The Hill model describes the mechanical properties of muscles and establishes the relationship between muscle activation and joint motion. In this study, personalized Hill models were customized for each participant, and the accuracy of the models was validated by comparing expected forces with actual force outputs. The experimental results show that under optimal conditions the normalized root mean square error between the measured force and the target force can be reduced to 5.1% of the maximum target force. Furthermore, this method enabled coordinated control of multiple fingers, facilitating a variety of grasping tasks. The proposed method offers significant potential for improving FES applications in hand rehabilitation and assistive robotics, providing a promising approach for precise force control in the rehabilitation of paralyzed patients.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"3594-3604"},"PeriodicalIF":5.2,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11150434","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144992219","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":"Assessment of Gait Pattern Changes in Lower Limb Amputees Using Inertial Sensor Signals: An Alternative to Gait Parameter Measurement","authors":"Gabriel Ng;Emilie Kuepper;Aliaa Gouda;Jan Andrysek","doi":"10.1109/TNSRE.2025.3605096","DOIUrl":"10.1109/TNSRE.2025.3605096","url":null,"abstract":"Effective gait monitoring and rehabilitation are essential for improving the quality of life in individuals with disabilities. Inertial sensors have the potential to enable long-term gait monitoring and assessment beyond the clinical setting. However, developing minimally intrusive systems that accommodate a wide range of gait deviations remains challenging. This study investigated an alternative to traditional approaches of using gait parameters for gait assessment, to evaluate whether changes in the overall gait patterns of lower-limb prosthetic users could be assessed by directly analyzing gyroscope and accelerometer data from inertial sensors. Eleven lower-limb prosthetic users completed walk trials with a biofeedback system designed to perturb gait patterns, while an additional twelve completed a gait training session with a physiotherapist. Inertial sensors were affixed at various locations along the lower body to collect gyroscope and accelerometer data. Three algorithms were evaluated: a hidden Markov model-based similarity measure (HMM-SM), self-organizing maps, and dynamic time warping. Statistical analyses demonstrated that self-organizing maps and dynamic time warping effectively assessed changes in gait patterns under a variety of gait perturbation strategies, with sensors located on the upper legs and lower legs significantly outperforming the pelvis location overall. The findings suggest the potential for wearable and adaptable gait monitoring systems capable of assessing changes in gait patterns. These systems could enable precise gait monitoring and real-time therapeutic intervention in real-world settings, offering a promising tool for long-term rehabilitation.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"3637-3646"},"PeriodicalIF":5.2,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11146927","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144952252","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":"Online Neural-to-Movement Mapping Transfer for Task Switching and Retention in Brain–Machine Interfaces","authors":"Zhiwei Song;Xiang Zhang;Mingdong Li;Jieyuan Tan;Yiwen Wang","doi":"10.1109/TNSRE.2025.3605246","DOIUrl":"10.1109/TNSRE.2025.3605246","url":null,"abstract":"Current brain-machine interfaces (BMIs) often rely on decoders trained for single tasks, limiting their flexibility in real-world applications. We propose an online learning framework that enables the transfer of neural-to-movement (knowledge) across tasks, supporting both task switching and memory retention. In our framework, neural activity is projected into a dynamical jPCA space to effectively dissociate into variant and invariant components. The variant components of the neural patterns are then aligned by deriving Gradient-based Kullback–Leibler Divergence Minimization (GKLD) for efficient online adaptation. A kernel reinforcement learning (KRL) model then decodes aligned neural signals while reusing prior neural-to-movement mapping. Evaluated on rats switching between a one-lever pressing and a two-lever discrimination task, the framework shows rapid convergence, over four times faster than the baseline method, and improves decoding accuracy by around 35% during task switching. Furthermore, when switching back to the original task, the framework successfully retains knowledge from the old task. Our method demonstrates general applicability to multiple task switching scenarios and maintains stable decoding across three representative days over a 21-day period, highlighting its potential for long-term, real-world use.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"3674-3684"},"PeriodicalIF":5.2,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11146924","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144952573","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":"Analysis of Freezing of Gait in Parkinson’s Disease Detection Using a Multimodal Prototype Learning Framework","authors":"Madhuri Thimmapuram;Ananda Rao Akepogu;P. Radhika Raju","doi":"10.1109/TNSRE.2025.3605204","DOIUrl":"10.1109/TNSRE.2025.3605204","url":null,"abstract":"Patients with severe Parkinson’s disease (PD) frequently have freezing of gait (FOG), a gait disability. By anticipating FOG before it occurs, pre-emptive cueing can either prevent FOG or lessen its severity and duration. To improve the accuracy of FOG detection, both electroencephalography (EEG) data and other complementary modalities, such as gait-based data, are increasingly being explored. The use of multimodal data is particularly important, as it enhances the robustness and accuracy of the detection models by combining different perspectives of the disease. Deep learning algorithms got a lot of attention in recent years for automated FOG identification; however their usefulness has been restricted due to a shortage of data samples, particularly medical data such as EEG. The scarcity of data can lead to overfitting in deep learning models, making it crucial for researchers to develop robust classification models that can operate effectively with a limited number of samples. Few-shot learning methods, such as prototype learning, have been introduced to mitigate these challenges by enabling models to effectively learn from a modest number of labeled samples. Thus in this research, we propose a prototype learning framework called CSE-ProtoNet, which utilizes CondenseNet with SEBlock for FOG detection in PD patients. Importantly, our study will leverage not only EEG data but also multimodal inputs, which can enhance the robustness and accuracy of PD detection. The method outperforms baseline models such as CSE-ProtoNet-ED, ProtoNet-CS, and ProtoNet-ED in terms of accuracy, F-score, recall, specificity, precision and AUC. The CSE-ProtoNet model also differentiated patients with FOG and Non-FOG with an accuracy of 98.75%. Cross-data validation was conducted to ensure the robustness and generalizability of the proposed method, confirming consistent performance across different folds.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"3709-3722"},"PeriodicalIF":5.2,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11146905","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144952582","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":"Assessing Cortical Excitability and Functional Network in Post-Stroke Unilateral Neglect: A Transcranial Magnetic Stimulation- Electroencephalography Investigation","authors":"Yichen Zhang;Linlin Ye;Wanying Zhao;Weiqun Song;Huanxin Xie;Lei Cao","doi":"10.1109/TNSRE.2025.3605150","DOIUrl":"10.1109/TNSRE.2025.3605150","url":null,"abstract":"Objective: Unilateral neglect (UN) is a common post-stroke condition, yet its underlying pathological mechanisms are not fully elucidated. This study aims to explore cortical function and brain network connectivity in patients with UN by employing transcranial magnetic stimulation-electroencephalography (TMS-EEG).Methods: The study involved three groups: 10 healthy controls, 10 patients with UN, and 10 stroke patients without UN. TMS was applied to the P4 site, and the resulting EEG responses were recorded and analysed. Results: UN patients demonstrated significant alterations in the N100 (p <0.05) and cortical excitability amplitude (CEA) between 25–275 ms (p <0.05). Time-frequency analysis further revealed a marked increase in theta-band activity in the parietal region of UN patients. Source localisation analysis indicated that, in later stages, neural signal propagation was predominantly confined to the parietal region. Time-varying EEG network analysis revealed that, during the early stages, patients with UN exhibited enhanced interhemispheric frontal connectivity, coupled with a reduction in connectivity between the affected parietal lobe and the contralateral frontal lobe. In the later stages, a notable decrease in parieto-occipital connectivity was observed, alongside increased connectivity within the frontal and parietal regions of both hemispheres, suggesting a widespread functional reorganisation of brain networks. Conclusion: UN patients exhibit diminished cortical excitability in the affected parietal cortex, coupled with altered brain network connectivity and increases in connectivity within specific regions. These findings suggest that the brain engages in functional compensation and network reorganisation to mitigate the neurological deficits induced by stroke.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"3583-3593"},"PeriodicalIF":5.2,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11146915","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144952628","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":"Association Between Cognitive Function and Spatial–Temporal Measures of Gait and Balance When Navigating a Virtual Reality Floor Maze","authors":"Jiawei Chen;Dario Martelli;Sunil K. Agrawal","doi":"10.1109/TNSRE.2025.3605536","DOIUrl":"10.1109/TNSRE.2025.3605536","url":null,"abstract":"Spatial navigation has been used as a behavioral marker of cognitive impairments. Floor Maze Tests (FMT) are used to characterize navigation where subjects physically move through a two-dimensional maze drawn on the floor. A Virtual Reality version of FMT (VR-FMT) has been developed, which provides a 3-dimensional navigation environment where the height of the maze walls can be altered. For both FMT and VR-FMT, the time used to complete the maze has been reported as the outcome measure to characterize the cognitive function. This study aims to show new performance metrics derived from spatial-temporal gait and balance parameters during navigation through the maze and their association with the cognitive scores in subjects with probable dementia. Sixty-five older adults with probable dementia participated in an experiment where subjects walked in VR-FMT with two wall heights, 2 centimeters (no wall condition) and 2 meters (wall condition). Our results showed that in no wall condition, the gait and balance parameters during navigation were associated with cognitive scores measuring attention and executive function. In wall condition, besides attention and executive function, gait parameters showed a correlation with the scores of the auditory memory. This paper showed that the spatial-temporal gait and balance parameters during spatial navigation are important metrics of cognitive function in addition to the completion time. VR-FMT with walls can help identify early memory impairments in individuals.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"3534-3543"},"PeriodicalIF":5.2,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11146926","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144952240","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":"Visual Feedback Gain Affects Cognitive Motor Function During Constant Grip Strength Control: A Functional Near-Infrared Spectroscopy Study","authors":"Yehong Zhang;Yinping Wei;Mengsheng Xia;Shiyang Lv;Wu Ren;Zongya Zhao;Ting Pang;Xuezhi Zhou;Yi Yu;Zhixian Gao","doi":"10.1109/TNSRE.2025.3605288","DOIUrl":"10.1109/TNSRE.2025.3605288","url":null,"abstract":"Visual feedback gain critically affects feedback quality and fine motor control, yet its neural basis related to cognitive motor control remains unclear. Nineteen healthy right-handed participants performed constant grip tracking at 20% of maximum voluntary contraction under low, medium, and high visual feedback gains. Functional near-infrared spectroscopy recorded hemodynamic responses from six regions of interest (ROIs): left/right prefrontal cortex (LPFC/RPFC), dorsolateral prefrontal cortex (DLPFC), left supplementary motor and premotor area (LSMA&PMA), left primary motor cortex (LM1), and left primary somatosensory cortex (LS1). Simultaneous grip force was collected. Under the medium gain level, the mean absolute error (MAE) was significantly lower than under both low (<inline-formula> <tex-math>${P}lt 0.001$ </tex-math></inline-formula>) and high (<inline-formula> <tex-math>${P}={0}.{036}$ </tex-math></inline-formula>) gain levels. Compared to the low gain level, the medium gain level showed higher HbO peaks in the RPFC (<inline-formula> <tex-math>${P}={0}.{022}$ </tex-math></inline-formula>), DLPFC (<inline-formula> <tex-math>${P}={0}.{011}$ </tex-math></inline-formula>), LSMA&PMA (<inline-formula> <tex-math>${P}={0}.{041}$ </tex-math></inline-formula>) and LS1 (<inline-formula> <tex-math>${P}={0}.{032}$ </tex-math></inline-formula>), greater phase-locking value between ROIs within the LPFC-RPFC (<inline-formula> <tex-math>${P}lt 0.001$ </tex-math></inline-formula>), RPFC-DLPFC (<inline-formula> <tex-math>${P}={0}.{047}$ </tex-math></inline-formula>) and DLPFC-LS1 (<inline-formula> <tex-math>${P}={0}.{030}$ </tex-math></inline-formula>), along with enhanced global coherence and higher clustering coefficients. Moreover, under the medium gain level, motor performance was significantly positively correlated with cortical activation across all six ROIs (<inline-formula> <tex-math>${P}lt 0.05$ </tex-math></inline-formula>). These findings suggest medium visual feedback gain optimally balances spatial information, enabling efficient neural resource allocation and enhanced motor performance. This study offers novel insights into the neural mechanisms of visually guided precision grip.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"3524-3533"},"PeriodicalIF":5.2,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11146931","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144952557","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":"Novel Physics-Informed Bayesian Fusion Post-Processor for Enhanced Gait Phase Recognition Using Surface Electromyography","authors":"Rami Mobarak;Alessandro Mengarelli;Rami N. Khushaba;Ali H. Al-Timemy;Federica Verdini;Sandro Fioretti;Laura Burattini;Andrea Tigrini","doi":"10.1109/TNSRE.2025.3604618","DOIUrl":"10.1109/TNSRE.2025.3604618","url":null,"abstract":"Myoelectric pattern recognition systems serve as a promising predictive control approach for the lower limbs prostheses and exoskeletons. However, their actual deployment is challenged by the signal stochastic nature that could contaminate the decision stream with physiologically implausible transitions, posing safety and metabolic cost concerns on the potential user. Therefore, this study proposes a novel Physics-Informed Bayesian Fusion (PI-BF) post-processor that embeds biomechanical sequentiality constraints into the posterior probabilistic output of the classifiers to suppress unstable transitions and promote natural gait progression. Time-Domain (TD) and Time-Dependent Power Spectrum Descriptors (TD-PSD) features were extracted from the lower limbs muscles surface electromyography (sEMG) signals and classified using Support vector machines (SVM), Artificial neural networks (ANN), K-Nearest Neighbour (KNN), and a CNN-LSTM hybrid deep learning model to predict five phases of gait cycle. The output of these classifiers was followed by the proposed PI-BF postprocessor and it was compared against Bayesian Fusion (BF) Majority voting (MV) as well as the performance without post-processing (WPP) using different numbers of votes from the previous windows. Results shows that PI-BF can increase the classification accuracy by up to 5.5% reaching up to 85% in SIAT-LLMD dataset (40 subjects) using SVM with 3 previous decision windows. It also reduced Transition Detection Difference (TDD) to 0.1 ± 59.8 ms and improved output stability by 5%, as measured by the Instability (INS) index. The proposedPI-BF exhibited consistent improvements in real-time gait phase recognition experiments, achieving classification accuracies of around 90%. These results demonstrate that PI-BF offers a practical, low-complexity solution for enhancing the safety, reliability, and real-time performance of myoelectric control in assistive lower-limb devices.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"3476-3487"},"PeriodicalIF":5.2,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11145891","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144952544","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":"Gamma-Band Binaural Beats Neuromodulation Enhances P300 Classification in an Auditory Brain-Computer Interface Paradigm","authors":"Xiaodong Li;Xin Wang;Shihao Chen;Wenxuan Zhu;Richu Jin;Weiwei Peng","doi":"10.1109/TNSRE.2025.3604016","DOIUrl":"10.1109/TNSRE.2025.3604016","url":null,"abstract":"While established neuromodulation techniques like transcranial magnetic stimulation and transcranial direct current stimulation have shown potential for enhancing brain-computer interface (BCI) performance, their clinical adoption faces challenges including high implementation costs, technical complexity, and safety concerns. This study investigated binaural beats (BB), a non-invasive auditory neuromodulation method characterized by operational simplicity and minimal adverse effects, as a practical alternative for optimizing auditory P300-BCI. Employing a crossover experimental design, thirty healthy participants underwent gamma-band (40 Hz) and alpha-band (10 Hz) BB stimulation in separate sessions. Auditory oddball paradigm experiments were conducted before and after each BB intervention. Electroencephalogram (EEG) data were decoded using both a machine learning classifier and a deep learning model for P300 classification. Additionally, irregular-resampling auto-spectral analysis (IRASA) was applied to extract aperiodic components from EEG during BB stimulation to evaluate changes in brain state. The results demonstrated frequency-dependent modulation effects: gamma-BB significantly improved P300 classification accuracy while alpha-BB impaired performance. Neurophysiological analysis revealed that gamma-BB decreased the aperiodic exponent, indicating enhanced brain arousal level, whereas alpha-BB produced the opposite pattern. Importantly, the aperiodic parameter change showed a significant association with BCI performance improvement. These findings established gamma-BB as an effective, low-cost neuromodulation strategy for augmenting auditory P300-BCI through brain state modulation.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"3455-3465"},"PeriodicalIF":5.2,"publicationDate":"2025-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11145223","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144952428","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}