@article {pmid41350592,
year = {2025},
author = {Gao, X and Lin, H and Wu, X and Zhang, D},
title = {Integrating active brain-computer interfaces (aBCIs) with passive BCIs (pBCIs) under different frustration levels.},
journal = {Scientific reports},
volume = {},
number = {},
pages = {},
doi = {10.1038/s41598-025-30168-1},
pmid = {41350592},
issn = {2045-2322},
abstract = {The mental state of the users can significantly affect the performance of active brain-computer interfaces (aBCIs). In this work, we aim to adopt passive BCIs (pBCIs) to measure a typical mental state, frustration, which is much relevant to aBCIs. A novel paradigm has been developed that combines both aBCIs and pBCIs under different frustration levels of users. The aBCI in this work is based on classic binary motor imagery (MI). In experiments, a new strategy was implemented that uses visual feedback to induce different levels of frustration. The electroencephalography (EEG) data collected were used for both aBCIs and pBCIs. The pBCI was utilized to assess the frustration level during the aBCI tasks, and the aBCI classification models for different levels of frustration were trained. For pBCI, the filter bank common spatial pattern (FBCSP) feature extraction and support vector machine (SVM) classification were utilized to classify three (i.e., low, moderate, high) frustration levels. For aBCI, the same method (FBCSP+SVM) was used to classify left versus right MI. We also aim to improve the performance of aBCIs in such conditions, so we developed two new methods to incorporate the pBCI results to adapt three MI classifiers to the varying states of frustration. Compared to the conventional approach of directly classifying MI tasks without considering frustration, the two proposed methods increased the mean classification accuracy by 7.40% and 8.62%, respectively. (Compared with the commonly used non-emotional discrimination data, the results are improved by 4.56% and 5.87% respectively.) Within the scope of non-invasive EEG and MI-based aBCI, this study provides, to our knowledge, an initial integrated demonstration in which a frustration-level classifier (pBCI) is trained and then used to adapt MI decoding (aBCI). It should not be taken as a claim of originality beyond this context. Starting from "user subjective perception", this paper rises to the engineering level of "objective frustration recognition and classification model adaptation", and makes a contribution to the depth of EEG data analysis and methodological integrity.},
}

@article {pmid41350471,
year = {2025},
author = {Yamaguchi, T and Hashimoto, RI and Sato, H},
title = {Cortical Representation of Auditory Selective Attention in a Dichotic Listening Task: A Functional Near-Infrared Spectroscopy Study.},
journal = {Brain topography},
volume = {39},
number = {1},
pages = {8},
pmid = {41350471},
issn = {1573-6792},
mesh = {Humans ; Spectroscopy, Near-Infrared/methods ; *Attention/physiology ; Male ; Female ; Dichotic Listening Tests ; Young Adult ; *Auditory Perception/physiology ; Adult ; Acoustic Stimulation ; Brain-Computer Interfaces ; Brain Mapping ; Music ; Reading ; *Cerebral Cortex/physiology ; },
abstract = {To advance the application of functional near-infrared spectroscopy (fNIRS) in brain-computer interface (BCI) technology, we investigated cortical activation patterns associated with auditory selective attention. Using a dichotic listening paradigm, participants were presented with simultaneous music and reading sounds to the left or right ear. During fNIRS recordings, they were instructed to selectively attend to the sound attribute (music vs. reading) or the spatial location (left vs. right ear). Cortical activity differences related to attentional targets were analyzed using a two-way analysis of variance (ANOVA), with sound attribute and spatial information as factors. Our results revealed a significant main effect of the sound attribute factor across multiple measurement channels. Notably, the right parietal region exhibited consistently greater activation when attention was directed toward music compared to reading sounds. Conversely, bilateral dorsolateral prefrontal cortex (DLPFC) channels showed higher activation when participants attended to reading sounds than to music. These findings indicate that cortical activation patterns are modulated by auditory attentional states based on sound attributes. Furthermore, preliminary classification analyses achieved an accuracy of 73.7% in discriminating attentional targets (music vs. reading sounds), demonstrating the feasibility of fNIRS-based BCI applications.},
}

Humans
Spectroscopy, Near-Infrared/methods
*Attention/physiology
Male
Female
Dichotic Listening Tests
Young Adult
*Auditory Perception/physiology
Adult
Acoustic Stimulation
Brain-Computer Interfaces
Brain Mapping
Music
Reading
*Cerebral Cortex/physiology

@article {pmid41350343,
year = {2025},
author = {Houmani, N and Yabouri, R and Garcia-Salicetti, S and Bedoin, M and Medani, T and Andrade, K},
title = {Individual neural dynamics of successful Gamma neuromodulation through EEG-neurofeedback in the aging brain.},
journal = {Scientific reports},
volume = {},
number = {},
pages = {},
doi = {10.1038/s41598-025-30212-0},
pmid = {41350343},
issn = {2045-2322},
abstract = {Gamma-band synchronization is a key mechanism for healthy cognitive function, yet it tends to decrease with age. EEG-based Neurofeedback (EEG-NF) is a promising tool enabling subjects to modulate their brain activity. However, its efficacy at the individual level remains unclear, which may partly explain the heterogeneity of neurofeedback outcomes. The primary objective of this study was to investigate individual neural dynamics of Gamma-band synchronization through EEG-NF training. We analyzed data from a double-blind, placebo-controlled trial using an EEG-based brain-computer interface, involving healthy older adults with subjective memory complaints, randomly assigned to a neurofeedback or a sham feedback group. Specifically, we employed a two-step unsupervised machine learning framework: first, epoch-based Agglomerative Hierarchical Clustering to identify individual-level response patterns, then Spectral Bi-Clustering to uncover higher-order structure at the population level. Results revealed a subgroup of individuals within the real neurofeedback condition who successfully enhanced their Gamma-band synchronization, with effects extending across the broader frequency spectrum. In contrast, the remaining participants in the neurofeedback group exhibited neural responses comparable to those observed in the sham group. This randomized controlled trial offers novel insights into the individual neural dynamics underlying successful Gamma EEG-NF training, highlighting its potential to promote healthy brain aging.},
}

@article {pmid41349820,
year = {2025},
author = {Solano-Suarez, KG and Arango-Sabogal, JC and Roy, JP and Molgat, E and Bédard, C and Gagnon, CA and Buczinski, S and Dufour, S},
title = {Bayesian diagnostic accuracy estimation of milk enzyme-linked immunosorbent assay, blood polymerase chain reaction, and peripheral blood lymphocyte count tests to determine bovine leukosis virus status in dairy cows.},
journal = {Journal of dairy science},
volume = {},
number = {},
pages = {},
doi = {10.3168/jds.2025-27485},
pmid = {41349820},
issn = {1525-3198},
abstract = {We assessed the diagnostic accuracy of an adapted antibody ELISA (ELISA-Ab) test, originally designed for bulk milk samples but applied on individual DHI-collected milk samples, to identify the bovine leukosis virus infection status of individual cows. Blood real-time PCR (qPCR) and blood lymphocyte count (LC) tests were used for comparison. For the milk ELISA-Ab, secondary objectives included identifying a fit-for-purpose threshold for result interpretation and evaluating whether the test's specificity could be influenced by the sampling technique (i.e., DHI-collected milk samples). Additionally, we evaluated whether the accuracy of each test varied with cow age, categorizing cows as young (2 to 4 yr old) or older (>4 yr old). In 2023, 8 dairy herds in Québec, Canada, were selected based on their historical within-herd leukosis prevalence, which was estimated to range from 10% to 75%. From all milking cows within these herds (n = 637), milk samples were collected during regular DHI, and blood samples were collected by the research team within one week of the DHI sampling. The indirect IDEXX Leukosis Milk Screening ELISA test was adapted to accommodate individual cow milk samples (as opposed to bulk tank milk samples), whereas an in-house qPCR assay targeting gag-pro-pol gene regions and LC determination were applied to blood samples. Bayesian latent class models were used to estimate the diagnostic accuracy of the tests. An optical density threshold of ≥0.5 for the ELISA-Ab provided an optimal control of the misclassification cost across various leukosis prevalence and, to a lesser extent, false negative to false positive cost ratio scenarios. With this threshold, the sensitivity and specificity estimates (95% Bayesian credible interval [BCI]) were 92% (BCI: 88%, 95%) and 99% (BCI: 96%, 100%), respectively. Sensitivity was higher in cows >4 yr old (99%, BCI: 96%, 100%) compared with cows 2 to 4 yr old (88%, BCI: 80%, 94%). We observed lower ELISA-Ab specificity in cows milked immediately after a positive cow (median: 82%, BCI: 72%, 97%) compared with those milked after a negative cow (median: 91%, BCI: 85%, 99%), suggesting a milk carryover effect due to the sampling technique. This carryover effect had a more pronounced impact on the false positive rate in herds with 30% to 50% leukosis prevalence, with the largest differences observed at the 30% prevalence scenario. However, the overall influence of the carryover effect remained limited. The qPCR test showed a sensitivity of 81% (BCI: 75%, 86%) and a specificity of 100% (98%, 100), whereas the LC test had a sensitivity of 55% (49%, 61%) and a specificity of 96% (93%, 98%). Both the qPCR and LC test accuracy parameters remained similar across age groups. In conclusion, the adapted ELISA-Ab test appears suitable for individual cow testing using DHI-collected milk samples, with higher sensitivity in cows >4 yr old. Its integration into existing milk recording programs provides a practical opportunity for herd-level leukosis monitoring.},
}

@article {pmid41349431,
year = {2025},
author = {Zou, T and Wang, X and Hu, X and Gao, Q and Han, H and Chen, H and Li, R},
title = {Distinct cortical morphometric inverse divergence changes in Parkinson's disease correlate with transcriptional expression patterns.},
journal = {NeuroImage. Clinical},
volume = {48},
number = {},
pages = {103916},
doi = {10.1016/j.nicl.2025.103916},
pmid = {41349431},
issn = {2213-1582},
abstract = {Growing evidence shows that parkinson's disease (PD) is a heterogeneous neurodegenerative disorder associated with region-specific changes in brain anatomy. However, the genetic mechanisms underlining these abnormalities are unclear. We aim to investigate PD neuroanatomical subtypes and uncover the specific brain-wide gene expression associated with morphometric abnormalities in each PD subtype. The morphometric inverse divergence (MIND) algorithm was used to quantify the morphological similarity based on multiple MRI features in 127 patients with PD and 101 healthy controls (HC). Then, heterogeneity through discriminant analysis (HYDRA) was employed to investigate the PD subtypes based on the MIND strength. Intergroup comparisons were conducted to assess MIND strength and clinical behavioral differences among PD subtypes and HC. Finally, we explored the associations between MIND network changes and gene expression in each PD subtype through partial least squares (PLS) regression, functional enrichment of PLS-weighted genes and transcriptional signature assessment of cell types. We identified two distinct subtypes of PD-related MIND changes, indicating that MIND decreased mainly in the frontal and cingulate cortices in subtype 1, and increased mainly in the occipital cortex and postcentral gyrus in subtype 2 (Bonferroni correction, p < 0.05). Both PD subtypes exhibited impaired cognitive function compared to HC, with subtype 2 showing lower Unified Parkinson's Disease Rating Scale Part III (UPDRS-III) and Hoehn and Yahr (H&Y) scores than subtype 1. Moreover, genetic commonalities analysis were identified 5 shared negative genes in the PD subtypes. Subtype 1 PLS1 genes were functionally enriched in biological processes related to synaptic function, neurodevelopment and degeneration. In addition, subtype 2 PLS1 genes showed additional involvement of metabolic pathways alongside synaptic function. Moreover, we identified MIND-related genes involved in inhibitory and excitatory neurons in subtype 1. In subtype 2, MIND-related genes were involved in astrocytes besides excitatory and inhibitory neurons. Our findings suggest two distinct neuroanatomical subtypes in PD, deepening the understanding of the heterogeneity of PD by bridging the gap between the transcriptome and neuroimaging.},
}

@article {pmid41348969,
year = {2025},
author = {Liu, X and Li, F and Czosnyka, M and Czosnyka, Z and Yu, H and Tong, X and Xing, Y and Li, H and Pu, K and Feng, K and Zhang, K and Pang, M and Ming, D},
title = {Multi-Omics and High-Spatial-Resolution Omics: Deciphering Complexity in Neurological Disorders.},
journal = {GigaScience},
volume = {},
number = {},
pages = {},
doi = {10.1093/gigascience/giaf137},
pmid = {41348969},
issn = {2047-217X},
abstract = {BACKGROUND: The world has witnessed a steady rise in neurological diseases, which represent a heterogeneous group of disorders characterized by complex pathogenesis involving disruptions at multiple molecular levels, including genomic, transcriptomic, proteomic, and metabolomic levels. These disorders, often caused by genetic mutations, metabolic imbalances, immune dysregulation, and environmental factors, pose significant challenges to global public health due to their high prevalence, mortality, and disability burden.

RESULTS: The advent of high-throughput technologies, such as next-generation sequencing and mass spectrometry, has provided valuable insights into the underlying mechanisms of disease, especially the development of multi- and high-spatial-resolution omics technologies, enabling the interaction of multiple levels of biology and analysis of the complex molecular networks and pathophysiological processes.

CONCLUSIONS: This review provides a comprehensive analysis of the latest advancements in multi- and high-spatial-resolution omics, with a focus on their applications in precision diagnostics, biomarker discovery, and therapeutic target identification in brain diseases. The study also highlights the current challenges in the clinical implementation and discusses the future directions, with artificial intelligence being anticipated to enhance clinical translation and diagnostic accuracy significantly.},
}

@article {pmid41348794,
year = {2025},
author = {Fu, Z and Zhang, P and He, X and Wang, H and Guo, Y and Chen, X and Huang, J},
title = {Deep Transfer Learning in Intra-subject and Inter-subjects for Intracortical Brain Machine Interface Decoding.},
journal = {IEEE transactions on bio-medical engineering},
volume = {PP},
number = {},
pages = {},
doi = {10.1109/TBME.2025.3640764},
pmid = {41348794},
issn = {1558-2531},
abstract = {OBJECTIVE: This study proposes an Improved Deep Transfer Network (IDTN) to enhance decoding accuracy, calibration efficiency, and adaptability of intracortical brain machine interface (iBMI) systems while reducing the reliance on new labeled samples.

METHODS: IDTN integrates two core components: Structural Joint Discriminative Maximum Mean Discrepancy (SJDMMD) and Kernel Norm Improved Multi-Gaussian Kernel (KNK). SJDMMD extends the standard MMD framework by incorporating a structure-enhanced soft label weighting mechanism that simultaneously minimizes intra-class distributional shifts and maximizes inter-class margins for precise cross-domain alignment. KNK employs multi-Gaussian kernels with kernel norm regularization to enhance high-dimensional feature representations and sharpen inter-class boundaries, thereby improving the effectiveness of SJDMMD.

RESULTS: Evaluated on neural datasets from two rhesus macaques, IDTN achieved superior performance in both intra- subject and inter-subject transfer scenarios, consistently outperforming state-of-the-art methods in decoding accuracy. IDTN also exhibited consistent decoding stability across daily recording sessions. Ablation studies further confirm that SJDMMD improves inter-class separability and intra-class coherence, while KNK contributes to more effective kernel mapping in complex feature spaces.

CONCLUSION: These findings underscore the effectiveness of structure-aware transfer learning for neural decoding.

SIGNIFICANCE: They also highlight the potential of IDTN for deployment in real-world iBMI applications, particularly in data-limited or cross-subject environments.},
}

@article {pmid41346965,
year = {2025},
author = {Mariscal, DM and Driscoll, B and Huang, H and Fisher, LE},
title = {Somatosensory restoration and neural control strategies in lower-limb prostheses.},
journal = {npj biomedical innovations},
volume = {2},
number = {1},
pages = {44},
pmid = {41346965},
issn = {3005-1444},
abstract = {People with lower-limb amputation cannot directly control or receive feedback from existing prostheses, but emerging technologies aim to address this gap. Some approaches focus on restoring somatosensation in the missing limb, while others record signals from residual muscles for prosthetic control. This review provides an overview of the current state of neuroprosthetics for somatosensory restoration and prosthetic control in lower-limb amputation, offering perspectives on integrating these technologies for bidirectional neuroprostheses.},
}

@article {pmid41346464,
year = {2025},
author = {Guragai, B and Jin, Z and Amos, TJ and Zhang, Q and Zhang, J and Li, L},
title = {Genetic contribution to intrinsic functional connectivity underlying general intelligence: evidence from adult twin study.},
journal = {Brain communications},
volume = {7},
number = {6},
pages = {fcaf461},
pmid = {41346464},
issn = {2632-1297},
abstract = {Resting-state functional connectivity has been linked to intelligence, and twin studies suggest that these associations may be influenced by genetic factors. To investigate this relationship, we analysed behavioural and resting-state functional magnetic resonance imaging data from young adult twins in the Human Connectome Project. General intelligence was assessed based on ten cognitive task performances. The results showed a positive correlation in both identical and fraternal twins, indicating a similarity of general intelligence among twin pairs. For the resting-state functional connectivity analysis, we conducted two approaches. In the first approach, twins were randomly assigned to two separate groups, ensuring that each pair was split between the groups. We then applied a connectome-based predictive method separately for identical and fraternal twins to predict general intelligence. Specifically, a predictive model was trained using one group's functional connectivity and then applied to its co-twin group to predict their general intelligence. Significant prediction was recorded in identical twins but not in fraternal twins, suggesting a high level of similarity of intelligence-related functional connectivity among identical twins. In the second approach, we aimed to quantify the intelligence similarity using the resting-state functional connectivity. To implement this, we generated models to predict the difference in general intelligence in twin pairs, where a smaller difference indicates a greater degree of similarity. The results showed that only the intelligence difference in identical twins was successfully predicted, where the default mode network showed a significant contribution, suggesting a higher neural basis for intelligence similarity in identical twins. Together, these findings demonstrate that functional connectivity patterns associated with intelligence extend across genetically identical twins. More broadly, they highlight the default mode network role in intelligence similarity and illustrate the utility of predictive modelling as a complementary framework to classical twin analyses.},
}

@article {pmid41282873,
year = {2025},
author = {Chen, H and Wang, J and Lai, S and Peng, G and Zong, G and Yuan, C and Luo, B},
title = {Smoking Cessation, Weight Change, and Risk of Dementia: A Prospective Cohort Study.},
journal = {medRxiv : the preprint server for health sciences},
volume = {},
number = {},
pages = {},
pmid = {41282873},
abstract = {OBJECTIVES: To assess the associations of smoking cessation and post-cessation weight gain with the risk of dementia and cognitive trajectories.

DESIGN: Prospective cohort study.

SETTING: The U.S. Health and Retirement Study (1995-2020).

PARTICIPANTS: A total of 32,802 dementia-free participants were included, with a mean age of 60.5 years (SD 10.7) and 57.1% female.

EXPOSURE: Smoking status and body weight were collected biennially via structural interviews.

MAIN OUTCOME MEASURES: Dementia was identified using the Langa-Weir algorithm. Cognitive function was assessed using a 27-unit scale. Cox proportional hazard models estimated hazard ratio (HR) of dementia by smoking cessation status, subsequent weight change, and duration of cessation. Among participants who quit during follow-up, linear mixed models assessed cognitive trajectories before and after cessation.

RESULTS: Over 25 years of follow-up, 5,868 dementia cases were documented. Compared with current smokers, individuals who quit during follow-up had a lower dementia risk after quitting (HR: 0.82, 95% confidence interval: 0.72-0.93), similar to those who had quit before baseline (0.76, 0.69-0.83) and to never smokers (0.72, 0.66-0.79). The benefits of cessation were largely limited to participants with no or modest weight gain (≤5 kg). By contrast, quitting accompanied by >10 kg weight gain was marginally associated with higher dementia risk (1.31, 0.95-1.80). Dementia risk declined steadily with increasing cessation duration, reaching the level of never smokers after approximately 5-7 years. Cognitive trajectory analyses showed that quitting was associated with long-term slower cognitive decline but no transient change, especially among those with no or modest weight gain.

CONCLUSIONS: Smoking cessation was associated with a sustained lower dementia risk and slower cognitive decline, comparable to benefits observed in never smokers and without evidence of a short-term risk increase. However, substantial post-cessation weight gain may attenuate these advantages. Smoking cessation programs should incorporate weight-management strategies to optimize long-term brain health.},
}

@article {pmid41345782,
year = {2025},
author = {Gebeyehu, TF and Sabbaghalvani, MA and Failla, G and Kabani, AS and Shah, Y and Kharichev, A and Dian, JA and Matsoukas, S and Vaccaro, AR and Schroeder, GD and Prasad, SK and Jallo, J and Heller, JE and Fehlings, MG and Harrop, JS},
title = {The application of artificial intelligence in the acute and sub-acute phases of spinal cord injury- a systematic review.},
journal = {Spinal cord},
volume = {},
number = {},
pages = {},
pmid = {41345782},
issn = {1476-5624},
abstract = {STUDY DESIGN: Systematic Review.

OBJECTIVE: To describe applications of AI for traumatic SCI management with focus on diagnostics, prognostication, and therapeutic interventions.

METHODS: PubMed, Scopus and Cochrane libraries were searched (March 2025). Studies published in English between January 1[st], 2020, and March 18, 2025, dealing with clinical aspects in the acute, post-injury rehabilitative and first year phases of SCI were included. Studies on brain computer interface, robotics and non-neurologic aspects of SCI were excluded. Extracted were country of study, study design, focus of study, total participants, American Spinal Injury Association (ASIA) Impairment Scale (AIS), machine learning (ML) models, inputs, outcomes and performance metrices.

RESULTS: A total of 23 studies with 120,931 individuals were identified. Classical Machine Learning Models, Ensemble Learning Models and Deep Learning Models were the most used ML families. Age, AIS, neurologic level of injury, sex, mechanism of injury and motor score were the most common inputs. Predictions of neurologic status, functionality status, Hospital/ICU utilizations, complications, survival, discharge destination and results of image segmentation and patient grouping were the outputs of interest. The performance metrices were satisfactory in most and higher than humans in some studies.

CONCLUSION: AI can facilitate personalized approach to diagnosis of SCI, prediction of outcomes like neurological improvement, complications, functionality indicators like walking, selfcare and independence, re-admissions, prolonged length of stays, discharge destination and mortality after injury. It was also useful to suggest specific MAP goals and time of surgical intervention. These functions complement clinical judgement.},
}

@article {pmid41345432,
year = {2025},
author = {Francis, N and Vadivu, G},
title = {ReHA-Net: a ReVIN-hybrid attention network with multiscale convolution for robust EEG artifact removal in brain-computer interfaces.},
journal = {Scientific reports},
volume = {},
number = {},
pages = {},
doi = {10.1038/s41598-025-28855-0},
pmid = {41345432},
issn = {2045-2322},
abstract = {Electroencephalography (EEG) is a non-invasive technique for monitoring brain activity, but its signal quality is frequently degraded by artifacts from ocular movements, muscle activity, and environmental noise. ReHA-Net is a deep learning framework for robust EEG denoising, combining a U-Net-based encoder-decoder with three core modules. (1) Hybrid Attention integrates temporal, spatial, and frequency attention to emphasize neural patterns while suppressing structured noise. (2) The Multiscale Separable Convolution (MSC) block employs dilated and parallel depth-wise separable convolutions with varying kernel sizes to capture both short-term and long-term temporal dependencies. (3) Reversible Instance Normalization (ReVIN) enhances cross-subject generalization while retaining subject-specific features. The model trains on an enhanced EEGdenoiseNet dataset with a wider signal-to-noise ratio range, combined artifact conditions, and tailored normalization strategies. ReHA-Net achieved strong denoising performance, with a PSNR of 27.10 dB, an SNR of about 17.06 dB, and a correlation coefficient of 0.976 with clean signals and a relative root mean square error (RRMSE) of 0.165. These outcomes demonstrate effective artifact reduction while maintaining neural activity, highlighting its suitability as a preprocessing step for tasks such as seizure detection, imagined speech decoding, and cognitive state monitoring.},
}

@article {pmid41345285,
year = {2025},
author = {Miao, T and Sha, L and Huang, K and Li, Y and Liu, B},
title = {SATrans-Net: Sparse Attention Transformer for EEG-based motor imagery decoding.},
journal = {Scientific reports},
volume = {},
number = {},
pages = {},
doi = {10.1038/s41598-025-30806-8},
pmid = {41345285},
issn = {2045-2322},
support = {SJC2022011//Basic Research Program of Suzhou/ ; SJC2022011//Basic Research Program of Suzhou/ ; SJC2022011//Basic Research Program of Suzhou/ ; SJC2022011//Basic Research Program of Suzhou/ ; SJC2022011//Basic Research Program of Suzhou/ ; BK20232008//Basic Research on Frontier Leading Technology in Jiangsu Province/ ; BK20232008//Basic Research on Frontier Leading Technology in Jiangsu Province/ ; BK20232008//Basic Research on Frontier Leading Technology in Jiangsu Province/ ; BK20232008//Basic Research on Frontier Leading Technology in Jiangsu Province/ ; BK20232008//Basic Research on Frontier Leading Technology in Jiangsu Province/ ; BE2021012-3 and BE2021012//Key Research and Development Program of Jiangsu/ ; BE2021012-3 and BE2021012//Key Research and Development Program of Jiangsu/ ; BE2021012-3 and BE2021012//Key Research and Development Program of Jiangsu/ ; BE2021012-3 and BE2021012//Key Research and Development Program of Jiangsu/ ; BE2021012-3 and BE2021012//Key Research and Development Program of Jiangsu/ ; },
abstract = {Brain-computer interface (BCI) technology decodes electroencephalography (EEG) signals to identify motor intentions associated with motor imagery (MI), offering assistive solutions for individuals with motor impairments. However, current deep learning methods often overlook the long-sequence nature of EEG-MI signals, leading to limited feature extraction and reduced decoding accuracy. To address this, we propose SATrans-Net, an end-to-end framework that models long-range dependencies in EEG-MI signals to enhance decoding performance. SATrans-Net uses two-dimensional depthwise separable convolution (2D-DSC) to extract spatiotemporal features and incorporates a Top-K Sparse Attention (TKSA) mechanism into the Transformer architecture, improving long-range modeling while reducing computational cost. By fusing local and global features, the model achieves accurate classification via a fully connected layer. For interpretability, Grad-CAM is applied to generate Class Activation Topography (CAT) maps, visualizing spatial attention over cortical regions. Cross-session evaluations show that SATrans-Net achieves average accuracies of 84.72%, 89.76%, and 96.79% on the BCI IV-2a, BCI IV-2b, and High-Gamma datasets, respectively, outperforming existing methods. Ablation studies further verify the critical role of the TKSA module. Overall, SATrans-Net demonstrates high decoding accuracy and strong interpretability, paving the way for the application of computational techniques in biomedical signal processing. Source Code:https://github.com/Jasmin-Tianhua/EEG-research_SATrans-Net.},
}

@article {pmid41345210,
year = {2025},
author = {Do, M and Evancho, A and Tyler, WJ},
title = {Bilateral transcutaneous auricular vagus nerve stimulation for the treatment of insomnia in breast cancer.},
journal = {Scientific reports},
volume = {},
number = {},
pages = {},
doi = {10.1038/s41598-025-30600-6},
pmid = {41345210},
issn = {2045-2322},
support = {PREP Award//University of Alabama at Birmingham/ ; },
abstract = {Substantial diagnostic and therapeutic advances have been made in medicine to address breast cancer. There remain unmet needs to translate solutions for addressing insomnia and mental health concerns in breast cancer patients. In this open-label, pilot clinical trial, we evaluated the safety and efficacy of nightly, bilateral, transcutaneous auricular vagus nerve stimulation (taVNS) on insomnia and mental health outcomes in breast cancer patients across a two-week treatment period. Our results demonstrate that noninvasive vagus nerve stimulation can significantly reduce insomnia severity, improve sleep quality, decrease sleep onset latency, and enhance sleep efficiency. Treatment with taVNS also significantly reduced the number of nightly awakenings, cancer-related fatigue, and depression scores while increasing heart rate variability. These observations demonstrate that auricular vagus nerve stimulation holds promise for improving sleep quality and mental health in patients diagnosed with breast cancer. Future investigations aimed at more thoroughly investigating the safety profile and clinical impacts of taVNS on the quality of life in patients with breast cancer are warranted.ClinicalTrials.gov Identifier: NCT06006299 23/08/2023.},
}

@article {pmid41345143,
year = {2025},
author = {Zhang, P and Yao, L and Yang, T and Lou, Y and Xu, W and Jiang, W and Li, W and Ji, X and Gao, F and Qian, Z},
title = {Revealing neural resonance in neuronal ensembles through frequency response tests.},
journal = {Scientific reports},
volume = {},
number = {},
pages = {},
doi = {10.1038/s41598-025-21252-7},
pmid = {41345143},
issn = {2045-2322},
support = {2024ZB661//Jiangsu Funding Program for Excellent Postdoctoral Talent/ ; ZKX24043//Key Project of Nanjing Health Science and Technology Development Special Fund/ ; BK20210531//Natural Science Foundation of Jiangsu Province/ ; NZ2024032//Fundamental Research Funds for the Central Universities/ ; 82151311//National Natural Science Foundation of China/ ; },
abstract = {Photobiomodulation emerges as a novel method to boost neuronal activities and brain function, with notable implications for treating brain disorders. Yet, the mechanisms and optimal frequency parameters of transcranial photobiomodulation are still unclear, which highlights a research gap in understanding how different stimulation frequencies affect neural responses. This study proposes a hypothesis that the nervous system exhibits resonance phenomena, suggesting that external stimuli near the system's resonant frequency trigger the strongest responses. We tested this by performing frequency response tests with pulsed transcranial near-infrared light (10-200 Hz) on mouse brains, monitoring neural responses across frequencies by analyzing cerebral blood flow, concentration of oxygenated hemoglobin, and neurophysiological activity in both cortical and deep brain regions. Our results reveal pronounced neural responses in cortical and deep brain areas at 60-80 Hz and 120-140 Hz, suggesting the potential existence of neural system resonance. Conceptually, the neural system appears to be modulatable by external stimuli, reaching maximal neural response when the stimulation frequency aligns with the system's resonant frequency, leading to neural resonance. These findings will expect to become guide new theoretical frameworks and strategies for neural modulation and therapeutic interventions.},
}

@article {pmid41344323,
year = {2025},
author = {Che, X and Zhao, H and Ye, X and Ye, S and Zhen, Z and Huang, Z and Li, Y and Zhang, S and Xu, P and Chen, X and Jiang, C and Pan, F and Luan, H and Chen, J and Shang, D and Hu, S and Tu, Y and Hu, L and Fitzgibbon, BM and Fitzgerald, PB and Cash, RFH and Huang, M},
title = {Frontoparietal network mediates the antidepressant effects of accelerated iTBS and cTBS: TMS-EEG study.},
journal = {Cell reports. Medicine},
volume = {},
number = {},
pages = {102470},
doi = {10.1016/j.xcrm.2025.102470},
pmid = {41344323},
issn = {2666-3791},
abstract = {Accelerated intermittent and continuous theta burst stimulation (a-iTBS and a-cTBS) show strong efficacy for treatment-resistant depression (TRD), yet their neural mechanisms remain unclear. This study uses concurrent transcranial magnetic stimulation (TMS) and electroencephalography (TMS-EEG) to examine these mechanisms in 40 TRD patients and 40 healthy controls (HCs). TRD individuals demonstrate abnormal local cortical excitability at baseline, characterized by left hypoactivity and right disinhibition. A-iTBS increases left excitability, and a-cTBS increases right inhibition, and both normalize it to the level of HCs. Network analyses reveal that a-iTBS improves current propagation to the left inferior parietal lobule (IPL), correlating with a better antidepressant effect. Contrastingly, a-cTBS induces a widespread inhibition as indicated by current propagation over parietal cortices, with the left IPL being most prominent, and this also correlates with a better antidepressant effect. These findings outline the frontoparietal circuitry in TMS antidepressant effects and provide insights for optimizing treatment efficacy. This study was registered at the Chinese Clinical Trial Registry (ChiCTR2200055320).},
}

@article {pmid41344290,
year = {2025},
author = {Liu, YJ and Wang, XD},
title = {Parallel supramammillary-hippocampal routes: Organization, dysregulation, and restoration.},
journal = {Neuron},
volume = {113},
number = {23},
pages = {3879-3881},
doi = {10.1016/j.neuron.2025.10.020},
pmid = {41344290},
issn = {1097-4199},
mesh = {Animals ; *Hippocampus/physiology ; Alzheimer Disease/physiopathology/pathology ; Neural Pathways/physiology ; Mice ; Humans ; },
abstract = {In this issue of Neuron, Luo et al.[1] report two supramammillary neuronal populations with segregated projections to the dorsal and ventral dentate gyrus that selectively modulate cognitive and emotional processes, respectively. Targeted activation of each pathway alleviates domain-specific behavioral deficits in an Alzheimer's disease mouse model.},
}

Animals
*Hippocampus/physiology
Alzheimer Disease/physiopathology/pathology
Neural Pathways/physiology
Mice
Humans

@article {pmid41341607,
year = {2025},
author = {Mahrouk, A},
title = {Symbolic feedback for transparent fault anticipation in neuroergonomic brain-machine interfaces.},
journal = {Frontiers in robotics and AI},
volume = {12},
number = {},
pages = {1656642},
pmid = {41341607},
issn = {2296-9144},
abstract = {BACKGROUND: Brain-Machine Interfaces (BMIs) increasingly mediate human interaction with assistive systems, yet remain sensitive to internal cognitive divergence. Subtle shifts in user intention-due to fatigue, overload, or schema conflict-may affect system reliability. While decoding accuracy has improved, most systems still lack mechanisms to communicate internal uncertainty or reasoning dynamics in real time.

OBJECTIVE: We present NECAP-Interaction, a neuro-symbolic architecture that explores the potential of symbolic feedback to support real-time human-AI alignment. The framework aims to improve neuroergonomic transparency by integrating symbolic trace generation into the BMI control pipeline.

METHODS: All evaluations were conducted using high-fidelity synthetic agents across three simulation tasks (motor control, visual attention, cognitive inhibition). NECAP-Interaction generates symbolic descriptors of epistemic shifts, supporting co-adaptive human-system communication. We report trace clarity, response latency, and symbolic coverage using structured replay analysis and interpretability metrics.

RESULTS: NECAP-Interaction anticipated behavioral divergence up to 2.3 ± 0.4 s before error onset and maintained over 90% symbolic trace interpretability across uncertainty tiers. In simulated overlays, symbolic feedback improved user comprehension of system states and reduced latency to trust collapse compared to baseline architectures (CNN, RNN).

CONCLUSION: Cognitive interpretability is not merely a technical concern-it is a design priority. By embedding symbolic introspection into BMI workflows, NECAP-Interaction supports user transparency and co-regulated interaction in cognitively demanding contexts. These findings contribute to the development of human-centered neurotechnologies where explainability is experienced in real time.},
}

@article {pmid41341241,
year = {2024},
author = {Kubben, P},
title = {Invasive Brain-Computer Interfaces: A Critical Assessment of Current Developments and Future Prospects.},
journal = {JMIR neurotechnology},
volume = {3},
number = {},
pages = {e60151},
pmid = {41341241},
issn = {2817-092X},
abstract = {Invasive brain-computer interfaces (BCIs) are gaining attention for their transformative potential in human-machine interaction. These devices, which connect directly to the brain, could revolutionize medical therapies and augmentative technologies. This viewpoint examines recent advancements, weighs benefits against risks, and explores ethical and regulatory considerations for the future of invasive BCIs.},
}

@article {pmid41338361,
year = {2025},
author = {Li, Y and Chen, S and Liu, YJ},
title = {Microglial phagoptosis in development, health, and disease.},
journal = {Neurobiology of disease},
volume = {},
number = {},
pages = {107211},
doi = {10.1016/j.nbd.2025.107211},
pmid = {41338361},
issn = {1095-953X},
abstract = {Microglial phagoptosis, defined as the phagocytosis of a viable cell by microglia that ultimately causes the death of the engulfed cell, has emerged as a pivotal process in sculpting neural circuits within the central nervous system (CNS). Essential for neurodevelopmental circuit refinement and ongoing tissue homeostasis, this process relies on dynamic molecular cues that direct microglia to specific cellular substrates. Physiologically, phagoptosis contributes to neural circuit refinement and cell number regulation during development; however, its dysregulation can drive neurodevelopmental and neurodegenerative disorders via aberrant cell removal. Recent advances have elucidated the distinct signaling pathways involved in target recognition and engulfment, revealing the dual roles of microglial phagoptosis in both CNS health and disease. Deeper mechanistic insight into this process offers new therapeutic opportunities for conditions characterized by defective or excessive cell clearance. This review summarizes current progress, highlights unresolved challenges, and discusses future perspectives on targeting microglial phagoptosis for intervention in CNS disorders.},
}

@article {pmid41337436,
year = {2025},
author = {Ding, Y and Wang, L and Wang, X and Chen, F},
title = {Developing Lightweight Models with Data Optimization for Attending Speaker Identity from EEG without Spatial Information.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11253106},
pmid = {41337436},
issn = {2694-0604},
mesh = {*Electroencephalography/methods ; Humans ; *Brain-Computer Interfaces ; Algorithms ; *Attention/physiology ; Signal Processing, Computer-Assisted ; Male ; Adult ; Artifacts ; },
abstract = {Spatial auditory attention decoding (Sp-AAD) holds great promise for brain-computer interfaces (BCIs). However, studies have shown that the high performance of Sp-AAD relies heavily on eye gaze artifacts rather than actual auditory attention features. For this reason, this study focuses on verifying whether EEG signals contain sufficient discriminative features for attending target speaker identity without eye gaze artifacts. In this study, we proposed an EEG-Mixup data optimization method to suppress trial-specific features in EEG data by adjusting the data distribution and generating soft labels through linear interpolation. In addition, a lightweight EEG-MLP model containing only 2.5k parameters was designed, which showed significant advantages over the latest SOTA model (DenseNet-3D) in cross-trial scenarios. It is shown that the model's generalization ability can be significantly improved by optimizing the data without increasing the data volume; meanwhile, the lightweight model demonstrates higher computational efficiency and inference speed in specific tasks. This study provides important theoretical and practical references for future optimization applications of BCI systems.Clinical Relevance- This study demonstrates the potential of lightweight EEG-based methods for attending target speaker identity without relying on eye gaze artifacts, providing a foundation for future auditory brain-computer interface systems.},
}

*Electroencephalography/methods
Humans
*Brain-Computer Interfaces
Algorithms
*Attention/physiology
Signal Processing, Computer-Assisted
Male
Adult
Artifacts

@article {pmid41337381,
year = {2025},
author = {Haqiqat, A and Karimi, N and Mirmahboub, B and Sobhaninia, Z and Shirani, S and Samavi, S},
title = {Tri-Model Integration: Advancing Breast Cancer Immunohistochemical Image Generation through Multi-Method Fusion.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-6},
doi = {10.1109/EMBC58623.2025.11252716},
pmid = {41337381},
issn = {2694-0604},
mesh = {Humans ; *Breast Neoplasms/diagnostic imaging/metabolism/diagnosis ; Female ; *Immunohistochemistry/methods ; *Image Processing, Computer-Assisted/methods ; Neural Networks, Computer ; Algorithms ; Reproducibility of Results ; },
abstract = {Immunohistochemical (IHC) staining is a crucial technique for diagnosing and formulating treatment plans for breast cancer, particularly by evaluating the expression of biomarkers like human epidermal growth factor receptor-2. However, the high cost and complexity of IHC staining procedures have driven research toward generating IHC-stained images directly from more readily available Hematoxylin and Eosin-stained images using image-to-image (I2I) translation methods. In this work, we propose a novel approach that combines the predictive capabilities of three state-of-the-art I2I models to enhance the quality and reliability of synthetic IHC images. Specifically, we designed a Convolutional Neural Network that takes as input a four-dimensional input comprising the outputs of three distinct models (each contributing an IHC prediction, which is an RGB three-dimensional output for each) and produces a final consensus image through a fusion mechanism. This ensemble method leverages the strengths of each model, leading to more robust and accurate IHC image generation. Extensive experiments on the BCI dataset demonstrate that our approach outperforms existing single-model methods, achieving superior Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) metrics. All of our code is available at: https://github.com/arshamhaq/BCI-fusion.Clinical RelevanceImproving the quality of synthetic IHC images can potentially reduce costs and streamline the diagnostic process, ultimately benefiting patient outcomes.},
}

Humans
*Breast Neoplasms/diagnostic imaging/metabolism/diagnosis
Female
*Immunohistochemistry/methods
*Image Processing, Computer-Assisted/methods
Neural Networks, Computer
Algorithms
Reproducibility of Results

@article {pmid41337376,
year = {2025},
author = {Kim, H and Ahn, M and Jun, SC},
title = {A Brain Switch for SSVEP-Based BCI Speller Using an RNN-Based Detection Approach.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-5},
doi = {10.1109/EMBC58623.2025.11252734},
pmid = {41337376},
issn = {2694-0604},
mesh = {Humans ; *Brain-Computer Interfaces ; *Evoked Potentials, Visual/physiology ; Electroencephalography/methods ; *Brain/physiology ; Algorithms ; Signal Processing, Computer-Assisted ; Male ; Adult ; *Neural Networks, Computer ; Female ; },
abstract = {Steady-state visual evoked potentials (SSVEP)-based brain-computer interface (BCI) systems are used commonly as spellers because they have high information transfer rate and high accuracy relative to other BCI paradigms. Asynchronous BCI systems allow users to input commands whenever they wish to use them, which may make these systems more realistic and practical than synchronous systems. In contrast, asynchronous BCIs, known as the Brain Switch, require robust mechanisms to detect users' intentions accurately while maintaining classification performance. This highlights the need for a BCI system that distinguishes users' intentions reliably. SSVEP paradigms often show variability in their frequency designs. In this study, we propose a two-stage asynchronous BCI system that combines a robust brain switch model that uses autocorrelation and Long Short-Term Memory (LSTM)) for detection and an EEGNet-based classifier. Our proposed system was evaluated using a 40-class SSVEP dataset involving 40 subjects. It achieved an impressive detection performance with a sensitivity (SEN) of 98.24 ± 2.21% and specificity (SPC) of 82.28 ± 11.63% for even 1-second epochs. Further, the system attained a classification accuracy (ACC) of 77.05 ± 14.95%. This model demonstrates significant potential to help develop more realistic and practical asynchronous BCI systems.},
}

Humans
*Brain-Computer Interfaces
*Evoked Potentials, Visual/physiology
Electroencephalography/methods
*Brain/physiology
Algorithms
Signal Processing, Computer-Assisted
Male
Adult
*Neural Networks, Computer
Female

@article {pmid41337322,
year = {2025},
author = {Zhao, R and Zhang, S and Bai, Y and Ni, G},
title = {Neural Dynamics in Imagined Speech: A Spatiotemporal Analysis Based on EEG Source Localization and Functional Connectivity.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-5},
doi = {10.1109/EMBC58623.2025.11254701},
pmid = {41337322},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; *Speech/physiology ; Male ; *Imagination/physiology ; Brain-Computer Interfaces ; Female ; Adult ; Spatio-Temporal Analysis ; *Brain/physiology ; Brain Mapping/methods ; Young Adult ; },
abstract = {Communication is a crucial part of daily life. However, patients with speech disorders may have difficulty communicating with the outside world and, in severe cases, may even completely lose the ability to speak. Imagined speech is an intrinsic speech activity that does not explicitly move any vocal organs, which has emerged as a promising avenue for brain-computer interface (BCI) research. In this study, we developed a novel experimental paradigm tailored to imagined speech tasks based on Chinese characters and collected participants' high-temporal-resolution electroencephalogram (EEG) data. Using dynamic statistical parametric mapping (dSPM), we delineated the spatial distribution of neural activation, while functional connectivity was quantified through phase-locking value (PLV) analysis to capture the temporal interplay between distinct brain regions. We introduced a novel spatiotemporal feature representation, termed information flow (IF), by segmenting the imagined speech process into 10 continuous temporal windows, we systematically analyzed the evolution of global and local information flow dynamics. The results revealed distinct spatiotemporal patterns of neural activation and functional connectivity, underscoring the coordinated interaction of critical brain regions involved in the process of imagined speech, which help to elucidate the spatiotemporal dynamics of imagined speech and provide valuable insights into its underlying neural mechanisms. This work provides a foundation for advancing speech BCI applications and contributes to understanding the cognitive and neural bases of imagined speech in Chinese.},
}

Humans
*Electroencephalography/methods
*Speech/physiology
Male
*Imagination/physiology
Brain-Computer Interfaces
Female
Adult
Spatio-Temporal Analysis
*Brain/physiology
Brain Mapping/methods
Young Adult

@article {pmid41337318,
year = {2025},
author = {Yadav, A and Garcia, FC and Gonzalez, A and Trevisan, BE and Xu, A and Ugur, M and Bhattacharjee, A and Pothukuchi, RP},
title = {Foresee: A Modular and Open Framework to Explore Integrated Processing on Brain-Computer Interfaces.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11254710},
pmid = {41337318},
issn = {2694-0604},
mesh = {*Brain-Computer Interfaces ; Humans ; Algorithms ; *Signal Processing, Computer-Assisted ; Electroencephalography ; },
abstract = {Brain-computer interfaces (BCIs) with processing integrated on the device enable fast and autonomous closed-loop interaction with the brain. While such BCIs are rapidly gaining traction, they are also difficult to design due to the tight and conflicting power and performance needs of on-device processing. Meeting these specifications often requires the BCI processors to be co-designed with applications and algorithms, with processor designers and computational neuroscientists working closely to converge on the target hardware platform. But, this process has traditionally been cumbersome and ad hoc, due to the lack of systematic design space exploration frameworks. In response, we present Foresee, a new framework for fast exploration of BCI processors. Foresee offers a unified and modular interface for iteratively co-optimizing BCI processors with their algorithms, without sacrificing accuracy, speed, or ease of use. Foresee is publicly available, and comes with a library of hardware blocks for common signal processing functions that the community could contribute and build on. We demonstrate Foresee's utility and capability by analyzing on-device processing for two seizure detection methods from prior work, and validating our analysis on real hardware. We expect Foresee to be vital in designing next-generation BCIs.},
}

*Brain-Computer Interfaces
Humans
Algorithms
*Signal Processing, Computer-Assisted
Electroencephalography

@article {pmid41337309,
year = {2025},
author = {Thapa, BR and Bae, J},
title = {A Window Analysis for the Decoding of Premovement and Movement Intentions in Freewill EEG.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11253481},
pmid = {41337309},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; Movement/physiology ; Male ; Female ; Support Vector Machine ; *Brain-Computer Interfaces ; Adult ; *Intention ; Young Adult ; Signal Processing, Computer-Assisted ; },
abstract = {Decoding movement-related intentions from electroencephalogram (EEG) is important for developing real-time brain machine interfaces (BMIs). While most studies focus on cue-based tasks in EEG-based BMIs, freewill reaching and grasping tasks allow subjects to initiate movements of their own will, making them relevant to practical EEG-based BMIs. However, the investigation of EEG window size for decoding freewill movements remains unexplored. This study systematically analyzes the effect of different window sizes on decoding EEG premovement (prior to the movement onset) and movement (after movement initiation) intentions in freewill reaching and grasping tasks. We used 49 EEG recordings from 23 subjects, and EEG windows of 0.1-1s in 0.1s increments were analyzed within the range of -3 to 3s relative to the movement onset at 0. Decoding was performed using regularized linear support vector machine (LSVM) and regularized linear discriminant analysis (RLDA), and performance was evaluated in terms of accuracy. Larger window sizes consistently outperformed smaller ones, with peak accuracy occurring between 0-1s relative to the movement onset. LSVM outperformed RLDA across all 10 window sizes, with peak accuracy ranging from 86.98% with 0.1s window to 90.94% with 1s window. Using LSVM, the earliest peak accuracy (90.03%) was achieved with a 0.7s window starting at 0.35s after the movement onset. Notably, a 0.5s window provided a peak accuracy of 89.5% which is not statistically significant compared to the 0.7s window (p = 0.05). The start point of the 0.5s window was 0.5s after the onset. With LSVM, considering the trade-off between decoding accuracy and latency, the 0.5s window offers the optimal choice for decoding movement intention in freewill EEG.Clinical relevance- Identifying the optimal window size to decode movement-related intentions in freewill EEG can help improve strategies to develop real-time BMIs for individuals with motor impairments.},
}

Humans
*Electroencephalography/methods
Movement/physiology
Male
Female
Support Vector Machine
*Brain-Computer Interfaces
Adult
*Intention
Young Adult
Signal Processing, Computer-Assisted

@article {pmid41337275,
year = {2025},
author = {Rutkowski, TM and Kasprzak, H and Otake-Matsuura, M and Komendzinski, T},
title = {Classifying Awareness with a Lightweight CNN in an Olfactory Oddball Passive BCI.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11253457},
pmid = {41337275},
issn = {2694-0604},
mesh = {Humans ; *Brain-Computer Interfaces ; Electroencephalography ; *Neural Networks, Computer ; *Awareness/physiology ; *Smell/physiology ; Algorithms ; Male ; Signal Processing, Computer-Assisted ; Adult ; Female ; },
abstract = {Olfaction, or the sense of smell, presents a promising avenue for enhancing brain-computer interface (BCI) usability and enabling passive cognitive state monitoring. In reactive BCI paradigms, odor cues can be associated with specific commands, facilitating more intuitive interaction. Furthermore, passive BCI applications can leverage olfactory stimuli to monitor cognitive processes. Despite this potential, challenges remain, notably the requirement for precise odor delivery mechanisms and robust algorithms capable of detecting and interpreting associated brain activity. This work proposes a novel approach, combining electroencephalography (EEG) and electrobulbogram (EBG) within an olfactory modality oddball paradigm, for predicting user awareness levels. A pilot study is presented, demonstrating improved user awareness classification performance with a newly developed multiclass, lightweight convolutional neural network (CNN) for this passive olfactory BCI modality, surpassing previously reported results.Clinical relevance- This research demonstrates the feasibility of inferring user awareness levels from concurrently acquired electroencephalographic (EEG) and electrobulbogram (EBG) neurophysiological data.},
}

Humans
*Brain-Computer Interfaces
Electroencephalography
*Neural Networks, Computer
*Awareness/physiology
*Smell/physiology
Algorithms
Male
Signal Processing, Computer-Assisted
Adult
Female

@article {pmid41337269,
year = {2025},
author = {Dijkema, EB and Pennartz, CMA and Olcese, U},
title = {A Proof-of-Concept Spike Based Neuromorphic Brain-Computer Interface.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11253485},
pmid = {41337269},
issn = {2694-0604},
mesh = {*Brain-Computer Interfaces ; Animals ; Mice ; Signal Processing, Computer-Assisted ; *Neural Networks, Computer ; *Action Potentials/physiology ; Visual Cortex/physiology ; },
abstract = {Closed-loop brain-computer interfaces (BCIs) hold promise for restoring function after neurological damage by dynamically processing neural signals and delivering targeted brain stimulation. To achieve clinically meaningful outcomes, such systems must operate with high spatiotemporal precision. This work aims to demonstrate a proof-of-concept neuromorphic BCI that processes neural spike events in near-real time, without necessitating preprocessing besides signal filtering and spike detection. Methods - We developed a system that acquires neural signals and streams spike events into a spiking neural network (SNN) running on SpiNNaker neuromorphic hardware. We evaluated the system's performance using both in vivo recordings from mouse visual cortex and simulated neural waveforms. We measured the roundtrip latency, defined as the time from spike detection to an output spike generated by the SNN. Results - Under baseline conditions with no hidden SNN layers, mean roundtrip latency was 4.69 ms (±1.70 ms). Adding hidden layers increased latency by approximately 3.65 ms per layer, reflecting the computational overhead of deeper networks. The system successfully detected and processed spikes in near real-time, demonstrating that neuromorphic hardware can manage spike-based input at speeds suitable for closed-loop intervention. Discussion - These findings indicate that neuromorphic SNNs can rapidly process neural signals, providing a foundation for closed-loop BCIs capable of bypassing damaged neural pathways. Future efforts will involve implementing stimulation protocols and functional SNNs. Such developments may ultimately facilitate more effective, flexible, and power-efficient neuroprosthetic devices.},
}

*Brain-Computer Interfaces
Animals
Mice
Signal Processing, Computer-Assisted
*Neural Networks, Computer
*Action Potentials/physiology
Visual Cortex/physiology

@article {pmid41337259,
year = {2025},
author = {Daling, MH and Alonzo, J and Lee, J and Lee, AH and Durfee, D and Larson, L and Nurmikko, A and Leung, VW},
title = {Shielded Relay Coil design to Optimize WPT and SAR for Distributed Wireless Brain Implants.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11253961},
pmid = {41337259},
issn = {2694-0604},
mesh = {*Wireless Technology/instrumentation ; Humans ; *Brain-Computer Interfaces ; Equipment Design ; *Brain/physiology ; *Prostheses and Implants ; },
abstract = {This paper presents a shielded relay antenna to simultaneously enhance Wireless Power Transfer (WPT) and reduce Specific Absorption Rate (SAR) for a network of distributed brain microimplants. Through strategic placement of conductive features, Eddy currents are created to oppose high magnetic fields. This design advantageously equalizes and increases the field strength over the cortical surface area. This work has the potential to address the WPT/ SAR co-optimization challenges for biomedical implants in general. When applied to the target 2 × 2 cm[2] wireless brain-machine interface (BMI) system operating at 915 MHz, HFSS simulations show it provides 1.2 dB WPT enhancement and a 29% SAR reduction.},
}

*Wireless Technology/instrumentation
Humans
*Brain-Computer Interfaces
Equipment Design
*Brain/physiology
*Prostheses and Implants

@article {pmid41337212,
year = {2025},
author = {Arjona, L and Rosenthal, J and Azkarate, M},
title = {Wireless Communication Protocol for backscatter-based Neural Implants.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11253936},
pmid = {41337212},
issn = {2694-0604},
mesh = {*Wireless Technology/instrumentation ; Humans ; *Brain-Computer Interfaces ; *Prostheses and Implants ; },
abstract = {This work presents a novel protocol for bidirectional wireless communication with neural implants that contributes to the growing field of closed-loop brain-computer interfaces (BCIs). BCIs are an emerging technology for studying and treating neurological disorders, such as spinal cord injuries. Furthermore, BCI heavily rely on neural implants as a crucial element, because they hold the potential to restore functionality of paralyzed limbs. The proposed protocol presents an open configuration to enable neural implants to communicate wirelessly with an external reader. Because computation to extract movement intention is performed externally, computing power is nearly unlimited and the energy consumption of the implant is reduced drastically. To validate the proposed protocol, the downlink (reader to implant) was implemented on a software defined radio running GNU-Radio toolkit with custom communication blocks. The uplink (implant to reader) was implemented on an FPGA. Finally, to validate the movement intention decoding, pre-recorded neural data was backscattered from an FPGA-based implant and the decoding was executed successfully.},
}

*Wireless Technology/instrumentation
Humans
*Brain-Computer Interfaces
*Prostheses and Implants

@article {pmid41337189,
year = {2025},
author = {Bleuze, A and Martel, F and Aksenova, T and Struber, L},
title = {Modification of cortical activation pattern after long-term BCI training and its impact on decoding model performances.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11253801},
pmid = {41337189},
issn = {2694-0604},
mesh = {*Brain-Computer Interfaces ; Humans ; *Electroencephalography/methods ; *Models, Neurological ; Male ; *Cerebral Cortex/physiology ; },
abstract = {In brain-computer-interfaces (BCIs) variability usually appears in brain signals from one session to another. This inter-session-variability is of major importance for two reasons. On the one hand it poses an issue for a model learned on previous session, that does not always perform correctly on new sessions. On the other hand, it can also be a marker of long-term adaptation in the brain of patients, which may reflect learning or even rehabilitation. This study investigates the phenomenon of physiological drift in BCIs, focusing on the evolution of brain activity over sessions. In order to do so, we analyzed the spatial patterns of synchronization and desynchronization in a wide range of frequencies. A linear regression model was proposed to quantify drift and residual variability. In this article, we study the inter-session variability both physiologically and from the point of view of the decoder performance and compute the correlation between them to examine their coherence. This study provides valuable insights on the physiological drift and its impact on BCI performance, contributing to the development of more stable and reliable BCI systems for rehabilitation medicine.(p)(p)Clinical Relevance-The long-term modifications in the activation patterns after BCI training studied in this article is an additional evidence of potential for rehabilitation using BCI.},
}

*Brain-Computer Interfaces
Humans
*Electroencephalography/methods
*Models, Neurological
Male
*Cerebral Cortex/physiology

@article {pmid41337178,
year = {2025},
author = {Wang, M and Wang, J and Zhao, J and Yao, L and Wang, Y},
title = {EIMNet: An EEG and iEEG-Fused Interactive Modality Network for Accurate Memory State Prediction during Working Memory Task.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-6},
doi = {10.1109/EMBC58623.2025.11253846},
pmid = {41337178},
issn = {2694-0604},
mesh = {Humans ; *Memory, Short-Term/physiology ; *Electroencephalography/methods ; *Brain-Computer Interfaces ; Algorithms ; Signal Processing, Computer-Assisted ; Male ; },
abstract = {Recent advancements in Brain-Computer Interface (BCI) research have increasingly highlighted the significance of multimodal integration for effectively extracting task-discriminative features. In the context of working memory (WM) task, we introduce EIMNet, a cross-modality fusion model inspired by the phase-amplitude coupling phenomenon. By enabling interaction between electroencephalography (EEG) and intracranial electroencephalography (iEEG), EIMNet enhances the representation of task-related features, improving the prediction of memory-related effects. Our ablation experiments demonstrate that EIMNet enhances decoding performance, with factors such as interaction factor selection, frequency band splitting, and data augmentation playing vital roles. We demonstrate the effectiveness of EIMNet in improving decoding accuracy by integrating EEG and iEEG for working memory task, with promising applications in memory and attention-related cognitive research.},
}

Humans
*Memory, Short-Term/physiology
*Electroencephalography/methods
*Brain-Computer Interfaces
Algorithms
Signal Processing, Computer-Assisted
Male

@article {pmid41337165,
year = {2025},
author = {Xu, Y and Otsuka, S and Nakagawa, S},
title = {Enhancing EEG-Based Emotion Classification by Refining the Spatial Precision of Brain Activity.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-6},
doi = {10.1109/EMBC58623.2025.11253823},
pmid = {41337165},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; *Emotions/physiology/classification ; *Brain/physiology ; Signal Processing, Computer-Assisted ; Neural Networks, Computer ; Brain-Computer Interfaces ; Algorithms ; },
abstract = {Advancements in neuroscience and deep learning have significantly enhanced bio-signal-based emotion recognition, a critical component in Brain-Machine Interface (BMI) applications for healthcare, human-computer interaction, and human-AI assistant communication. Former studies have proposed Manual Mapping electrode matrices and employing Convolutional Neural Networks (CNNs) to recognize spatial EEG activities. However, this Manual Mapping of EEG electrodes onto matrix grids limits spatial precision and introduces inefficiencies. This study proposes automated channel mapping methods of Orthographic Projection and Stereographic Projection to address these challenges, using Differential Entropy and Power Spectral Density with Linear Dynamical Systems as features. A 3-branch multiscale CNN was trained on open-source dataset, employing a 5-fold cross-classification approach. Experimental results demonstrate that higher-resolution grids (16×16, 24×24) with automated projections significantly outperform Manual Mappings, achieving up to a 4.06% improvement in classification accuracy (p < 0.05). This result indicates that enhancing spatial precision of EEG data improves emotion classification, establishing automated spatial mapping as an advancement in EEG-based emotion recognition.Clinical Relevance-Advancement in emotion classification accuracy can facilitate more reliable diagnostic tools and personalized therapeutic interventions for mental health disorders, such as depression and anxiety.},
}

Humans
*Electroencephalography/methods
*Emotions/physiology/classification
*Brain/physiology
Signal Processing, Computer-Assisted
Neural Networks, Computer
Brain-Computer Interfaces
Algorithms

@article {pmid41337115,
year = {2025},
author = {Rivelli, F and Popov, M and Kouzinopoulos, CS and Tang, G},
title = {Adaptively Pruned Spiking Neural Networks for Energy-Efficient Intracortical Neural Decoding.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11254088},
pmid = {41337115},
issn = {2694-0604},
mesh = {Animals ; *Brain-Computer Interfaces ; Algorithms ; *Neural Networks, Computer ; Humans ; *Neurons/physiology ; *Action Potentials/physiology ; *Nerve Net/physiology ; },
abstract = {Intracortical brain-machine interfaces demand low-latency, energy-efficient solutions for neural decoding. Spiking Neural Networks (SNNs) deployed on neuromorphic hardware have demonstrated remarkable efficiency in neural decoding by leveraging sparse binary activations and efficient spatiotemporal processing. However, reducing the computational cost of SNNs remains a critical challenge for developing ultra-efficient intracortical neural implants. In this work, we introduce a novel adaptive pruning algorithm specifically designed for SNNs with high activation sparsity, targeting intracortical neural decoding. Our method dynamically adjusts pruning decisions and employs a rollback mechanism to selectively eliminate redundant synaptic connections without compromising decoding accuracy. Experimental evaluation on the NeuroBench Non-Human Primate (NHP) Motor Prediction benchmark shows that our pruned network achieves performance comparable to dense networks, with a maximum tenfold improvement in efficiency. Moreover, hardware simulation on the neuromorphic processor reveals that the pruned network operates at sub-μW power levels, underscoring its potential for energy-constrained neural implants. These results underscore the promise of our approach for advancing energy-efficient intracortical brain-machine interfaces with low-overhead on-device intelligence.},
}

Animals
*Brain-Computer Interfaces
Algorithms
*Neural Networks, Computer
Humans
*Neurons/physiology
*Action Potentials/physiology
*Nerve Net/physiology

@article {pmid41337108,
year = {2025},
author = {Song, Q and Kang, G},
title = {A Multi-Band Self-Attention Network for Motor Imagery Classification.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11254113},
pmid = {41337108},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; *Brain-Computer Interfaces ; *Neural Networks, Computer ; Signal Processing, Computer-Assisted ; *Imagination/physiology ; Algorithms ; },
abstract = {Brain-computer interface (BCI) systems create a novel communication method between humans and machines by translating human thoughts into actionable commands to control external devices. Motor imagery (MI) electroencephalogram (EEG) signals have significant applicability in various medical and non-medical industries, including stroke rehabilitation, wheelchair control, and drone operation. However, the practical application of EEG remains limited by the decoding performance and generalization ability of MI signalsThis study introduces a multi-branch self-attention network for motor imagery (MI) signal classification. Each branch independently processes EEG signals decomposed into distinct frequency bands through convolutional neural networks (CNNs) and multi-head self-attention (MHA) mechanisms, enabling the extraction of both fundamental and discriminative spatial-temporal features. To further capture dynamic temporal dependencies, long short-term memory (LSTM) networks are integrated. We systematically evaluate three signal decomposition ensemble empirical mode decomposition (EEMD), wavelet packet decomposition (WPD), and brain rhythm-based decomposition-to optimize feature representation. Extensive experiments on the BCI Competition IV 2a dataset demonstrate state-of-the-art performance, with subject-dependent and subject-independent accuracies of 84.04% and 71.67%, respectively. Comparative analyses against benchmark models (EEGNet, EEGTCNet, ShallowConvNet, etc.) validate the superiority of our approach in classification accuracy and generalization capabilityClinical relevance- This study investigates the methods for decoding motor imagery EEG signals and establishes the positive role of each module in classification. The improvement in accuracy can lead to better outcomes in medical applications such as controlling prosthetics, wheelchairs, and stroke rehabilitation.},
}

Humans
*Electroencephalography/methods
*Brain-Computer Interfaces
*Neural Networks, Computer
Signal Processing, Computer-Assisted
*Imagination/physiology
Algorithms

@article {pmid41337106,
year = {2025},
author = {Zhong, Y and Wen, H and Assam, M and Yao, L and Wang, Y},
title = {Motor-Sensory Coupled Learning for Motor Imagery Decoding.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-5},
doi = {10.1109/EMBC58623.2025.11254055},
pmid = {41337106},
issn = {2694-0604},
mesh = {Humans ; *Brain-Computer Interfaces ; Electroencephalography ; *Imagination/physiology ; Stroke Rehabilitation ; Signal Processing, Computer-Assisted ; *Learning ; Male ; },
abstract = {Brain-Computer Interface (BCI) technology has significant potential for advancing stroke rehabilitation by promoting motor recovery by decoding motor intentions from electroencephalogram (EEG) signals. However, the practical application of BCI in rehabilitation faces several challenges, particularly in decoding accuracy. This limitation often stems from an overemphasis on motor imagery signals, while sensory components, which are crucial for effective motor function recovery, are frequently overlooked. In this paper, we propose a novel framework to enhance BCI performance by integrating both sensory and motor modalities through a motor-sensory coupled learning approach. The model leverages EEG data induced by both motor imagery (MI) and tactile sensation (TS), using adversarial training to capture the coupled features of these two domains. By incorporating reliable sensory signals, the proposed approach aims to improve the robustness and accuracy of motor imagery decoding, offering particular benefits for stroke patients with impaired motor rhythms. Experimental results from BCI-naive subjects show a significant improvement in classification accuracy compared to traditional motor imagery-only models, suggesting that this approach holds promise as a potential solution for stroke rehabilitation. These findings indicate that integrating sensory signals into BCI systems could lead to more effective rehabilitation strategies, paving the way for the development of more robust and adaptive BCI technologies in the future.},
}

Humans
*Brain-Computer Interfaces
Electroencephalography
*Imagination/physiology
Stroke Rehabilitation
Signal Processing, Computer-Assisted
*Learning
Male

@article {pmid41337085,
year = {2025},
author = {Ong, JX and Premchand, B and Lim, RY and Chew, E and Jiang, M and Tang, N and Ang, KK},
title = {Inhibitory Effects of Individualized Transcranial Alternating Current Stimulation on Motor Imagery and Interhemispheric Symmetry: Implications for Stroke Rehabilitation.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11254631},
pmid = {41337085},
issn = {2694-0604},
mesh = {Humans ; *Stroke Rehabilitation ; *Transcranial Direct Current Stimulation/methods ; Male ; Female ; Stroke/physiopathology ; *Imagination/physiology ; Adult ; Brain-Computer Interfaces ; },
abstract = {Transcranial alternating current stimulation (tACS) holds potential in stroke rehabilitation, but its effects when delivered at an individual's peak motor imagery (MI) frequency remain unclear. This study investigated the impact of tACS delivered at subject-specific peak MI frequencies on MI performance accuracy, quantified in terms of classification accuracy, and interhemispheric symmetry, measured via the brain symmetry index (BSI). Using a brain-computer-brain closed-loop system, each subject's peak MI performance frequency was first identified during the Pre-stimulation phase, after which tACS was delivered at this determined frequency. Our findings show that active individualized tACS decreased MI performance and increased BSI, suggesting inhibitory effects on motor-related neural processes.Clinical Relevance- The observed inhibitory effects of tACS highlight its potential for targeted neuromodulation in stroke recovery. Future research should explore how inhibitory effects can be harnessed therapeutically and investigate stimulation parameters that could optimize outcomes for functional recovery. The demonstrated ability of tACS to modulate brain activity, evidenced by increased BSI, underscores its promise as a neuromodulatory tool in clinical applications.},
}

Humans
*Stroke Rehabilitation
*Transcranial Direct Current Stimulation/methods
Male
Female
Stroke/physiopathology
*Imagination/physiology
Adult
Brain-Computer Interfaces

@article {pmid41337074,
year = {2025},
author = {Carvallo, A and Struber, L and Costecalde, T and Souriau, R and Charvet, G and Aksenova, T},
title = {Decoding of Individual Fingers Attempted Movement from Epidural ECoG in a Patient with Tetraplegia.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11254592},
pmid = {41337074},
issn = {2694-0604},
mesh = {Humans ; *Quadriplegia/physiopathology ; *Brain-Computer Interfaces ; *Fingers/physiopathology/physiology ; Movement/physiology ; Algorithms ; *Electrocorticography/methods ; Male ; },
abstract = {Brain-Computer interfaces (BCIs) enable direct communication between the brain and external devices. This technology holds significant potential for restoring motor function in individuals with severe neurological impairments. Among others, restoration of fine hand motor functions allowing grasping and objects manipulation is a priority for enhancing patients' lifestyle. Decoding finger movements is crucial for the precise control of hand neuroprosthetics. In this article, we analyzed neural activity of a tetraplegic patient implanted with two WIMAGINE ECoG recording devices in front of the sensorimotor cortex of both hemispheres. ECoG was recorded over three sessions while the patient attempted to move individual fingers on the right hand. The attempted finger movements was decoded using a Hidden Markov Model, integrating Recursive Sample Weighted - N-Ways Partial Least Square algorithm addressing class imbalance. In the offline study, we obtained balanced accuracy 0.6603 ± 0.0087 in average for decoding activation of five individual fingers. Our results shows that decoding individual fingers movements attempts is possible in ECoG, paving the way for fine movement restoration using BCI.Clinical Relevance- Efficient decoding of individual fingers attempted movements using chronic ECoG recording devices in a tetraplegic patient, suggesting the feasibility of hand neuroprosthesis aimed at fine hand motor restoration in impaired individuals.},
}

Humans
*Quadriplegia/physiopathology
*Brain-Computer Interfaces
*Fingers/physiopathology/physiology
Movement/physiology
Algorithms
*Electrocorticography/methods
Male

@article {pmid41337062,
year = {2025},
author = {Zhu, Z and Han, J and Zhang, Z and Wannawas, N and Faisal, AA},
title = {Identifying the Nature of Grip Force Signals in EEG & fNIRS with Multi-Modal Graph Fusion Network.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11254624},
pmid = {41337062},
issn = {2694-0604},
mesh = {Humans ; *Hand Strength/physiology ; *Electroencephalography/methods ; Brain-Computer Interfaces ; Spectroscopy, Near-Infrared/methods ; Male ; Adult ; Signal Processing, Computer-Assisted ; Female ; },
abstract = {Brain-Computer interfaces can assist motor rehabilitation for people with severe paralysis by directly decoding their brain signals into movement intention and executing with external devices without passing the impaired neural pathways. It is crucial to restore natural and smooth daily movements, and continuous force control is one of the most important kinaesthetic functions. However, the complex continuous force decoding and limited relevant public datasets greatly challenge this field. How the brain coordinates the motor command or sensory feedback during the force control behaviour also remains to be discussed. This work investigated these questions through a novel experimental setup by isolating the motor intention and sensory feedback and combining both components flexibly for hand grip. We applied functional electrical stimulation to induce passive gripping and collected grip force with multi-modal brain signals. Significant neural pattern differences were found in EEG time-frequency representation by comparing the brain responses under different task conditions, including voluntary movement, motor imagery, and passive perception status. Additionally, we present a multi-modal graph fusion model fusing both EEG and fNIRS for continuous bimanual grip force decoding. These contributions are beneficial to developing neural interfaces for rehabilitation and assistive devices that involve force manipulation or operate in isometric schemes.},
}

Humans
*Hand Strength/physiology
*Electroencephalography/methods
Brain-Computer Interfaces
Spectroscopy, Near-Infrared/methods
Male
Adult
Signal Processing, Computer-Assisted
Female

@article {pmid41337056,
year = {2025},
author = {Abdo, EA and Yakovlev, A and Degenaar, P},
title = {Multipolar Hybrid Stimulation for Visual Prostheses: Enhancing Resolution and Specificity.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11254606},
pmid = {41337056},
issn = {2694-0604},
mesh = {*Visual Prosthesis ; Humans ; *Electric Stimulation/methods ; Optogenetics ; Brain-Computer Interfaces ; *Visual Cortex/physiology ; Animals ; },
abstract = {Advancements in neural stimulation techniques are essential for improving the precision and efficiency of brain-machine interfaces, particularly in visual cortical prostheses. These prostheses aim to restore vision by stimulating the visual cortex, but current methods face challenges such as limited spatial resolution, high power consumption, and non-specific activation. This work proposes a multipolar hybrid stimulation approach that combines electrical and optical neuromodulation to mitigate these limitations. Unlike traditional monopolar and bipolar methods, which require numerous electrodes or suffer from crosstalk and timing issues, the proposed system employs polarity switching and selective electrode control, enabling customizable electric fields alongside optogenetics for precise neural targeting and enhanced resolution. By utilizing subthreshold electrical and optogenetic stimulation, this approach improves spatial selectivity, minimizes crosstalk, and reduces power consumption. The conceptual design for neural tissue stimulation is presented, with ongoing efforts focused on integrating this system into a microelectronic chip. By addressing key limitations in current prosthetic systems, this work contributes to the development of more efficient and scalable solutions for visual restoration.},
}

*Visual Prosthesis
Humans
*Electric Stimulation/methods
Optogenetics
Brain-Computer Interfaces
*Visual Cortex/physiology
Animals

@article {pmid41337025,
year = {2025},
author = {Liu, G and Yan, Y and He, S and Cai, J and Cheok, AD and Qi Wu, E and Song, A},
title = {A Neuromorphic Approach for Brain-Machine Interface Using Spiking Neural Networks.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11254255},
pmid = {41337025},
issn = {2694-0604},
mesh = {*Brain-Computer Interfaces ; Animals ; *Neural Networks, Computer ; Algorithms ; *Action Potentials/physiology ; Motor Cortex/physiology ; Macaca mulatta ; Humans ; },
abstract = {Brain-machine interfaces (BMIs) have emerged as a promising technology for restoring motor function in paralyzed individuals through direct neural control of prosthetic devices. While conventional decoding algorithms have achieved considerable success, they often overlook the fundamental biological properties of neural information processing. This paper presents a novel approach using Spiking Neural Networks (SNNs), a neuromorphic computing paradigm that closely mimics biological neural dynamics through event-driven processing and spike-timing-dependent plasticity. A SNN-based decoder was implemented for offline decoding of intracortical neural recordings from the primary motor cortex (M1) and dorsal premotor cortex (PMd) to continuous 2D cursor movements in a macaque monkey. This approach leverages the temporal processing capabilities of SNNs to capture the complex, time-varying nature of neural representations, potentially enabling more naturalistic and adaptive BMI control.},
}

*Brain-Computer Interfaces
Animals
*Neural Networks, Computer
Algorithms
*Action Potentials/physiology
Motor Cortex/physiology
Macaca mulatta
Humans

@article {pmid41337011,
year = {2025},
author = {Yao, R and Du, Z and Liang, F and Li, W and Hong, B},
title = {Dual-layer hand gestures decoding with wireless epidural braincomputer interface in a tetraplegia.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-6},
doi = {10.1109/EMBC58623.2025.11254206},
pmid = {41337011},
issn = {2694-0604},
mesh = {Humans ; *Brain-Computer Interfaces ; *Quadriplegia/physiopathology ; *Gestures ; *Hand/physiopathology/physiology ; *Wireless Technology ; Algorithms ; Electroencephalography ; },
abstract = {Spinal cord injury disrupts the neural connections between the brain and limbs, resulting in tetraplegia. Brain-computer interface (BCI) hold promise for enabling voluntary limb movements in tetraplegic patients, yet achieving fine motor control of the hand remains a challenge. Invasive BCI based on intracortical electrode arrays have demonstrated real-time multi-gesture decoding. However, their long-term safety is a major barrier in clinical applications. In this study, a tetraplegic patient was implanted with our recently developed wireless minimally invasive BCI, which records epidural field potential from eight electrodes over the sensorimotor cortex to decode continuous hand movement intentions. Natural hand movements can be decomposed into dual layers: the high level movement states and the low level finger kinematics. Accordingly, we propose a dual-layer decoding algorithm for multi-gesture BCI decoding. The upper layer infers the movement state using a hidden Markov model, while the lower layer decodes finger motion variables through a mixture of experts and filters them with a state specific linear system. This approach enables the real-time decoding of six hand gestures, outperforming classical decoders and recurrent neural networks. The proposed dual-layer framework achieves multi-gesture decoding solely from epidural EEG signals, paving the way for the development of flexible and robust BCI control of hand movement.},
}

Humans
*Brain-Computer Interfaces
*Quadriplegia/physiopathology
*Gestures
*Hand/physiopathology/physiology
*Wireless Technology
Algorithms
Electroencephalography

@article {pmid41336923,
year = {2025},
author = {Chen, X and Peng, Y and Li, C and Pan, Y and Ding, N and Zhang, S},
title = {MI-LTN: A Neurosymbolic Framework for Enhanced EEG Feature Extraction and Model Interpretability in MI-BCI.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11252655},
pmid = {41336923},
issn = {2694-0604},
mesh = {*Electroencephalography/methods ; *Brain-Computer Interfaces ; Humans ; *Signal Processing, Computer-Assisted ; Algorithms ; },
abstract = {Brain-Computer Interface (BCI) is a cutting-edge technology that facilitates human-computer interaction. Motor Imagery Electroencephalogram (MI-EEG) decoding technology has emerged as a significant direction in BCI research. Despite the remarkable advancements in deep learning for EEG signal decoding in recent years, two major challenges persist: the comprehensive representation and extraction of features, and the lack of interpretability. To address these issues, we propose a novel neurosymbolic framework termed MI-LTN (Motor Imagery Logic Tensor Network), incorporate logical constraints into the training model using the Logic Tensor Network (LTN) and employ Shapley values to evaluate and adjust the importance of channels. Our experimental results show that MI-LTN achieves classification accuracies of 86.00% and 88.84% on the BCI IV 2a and BCI IV 2b datasets, respectively. These results demonstrate the great potential of LTN in MI-EEG decoding.},
}

*Electroencephalography/methods
*Brain-Computer Interfaces
Humans
*Signal Processing, Computer-Assisted
Algorithms

@article {pmid41336899,
year = {2025},
author = {Bradshaw Bernacchi, JK and Lopez Valdes, A},
title = {Electrophysiological Characterisation of Commercial Ear-EEG Devices.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11252639},
pmid = {41336899},
issn = {2694-0604},
mesh = {*Electroencephalography/instrumentation ; Humans ; *Ear/physiology ; Adult ; Male ; Female ; Wearable Electronic Devices ; Electrodes ; Brain-Computer Interfaces ; Equipment Design ; Young Adult ; Signal Processing, Computer-Assisted ; },
abstract = {Ear-EEG devices are advanced wearables revolutionizing EEG technology by combining comfort and portability. With the increasing availability of commercial ear-EEG devices, there is a need for an independent characterisation of the electrophysiological performance to guide users and researchers. Here, we evaluate the performance of the IDUN Guardian Earbuds (IGEB, IDUN Technologies AG) by analysing electrophysiological responses to several well-established EEG paradigms, including event-related potentials (ERPs), auditory steady-state response (ASSR), steady-state visually evoked potential (SSVEP), and alpha block, and comparing them to standard scalp-based EEG recordings acquired simultaneously from eight participants utilizing a validation toolkit previously developed in our lab. Results indicate that the in-ear device is capable of detecting SSVEPs. However, we did not observe ERPs, ASSRs, or alpha blocking. Simulating in-ear EEG with electrode T8 referenced to T7 slightly improved the quality of the signal, which was further enhanced with midline reference electrodes.Clinical Relevance- Characterising this technology marks a step forward providing independent assessment of commercially available devices in view of expanding EEG applications, from long-term monitoring and wearable health solutions to advanced brain-machine interfaces (BMI).},
}

*Electroencephalography/instrumentation
Humans
*Ear/physiology
Adult
Male
Female
Wearable Electronic Devices
Electrodes
Brain-Computer Interfaces
Equipment Design
Young Adult
Signal Processing, Computer-Assisted

@article {pmid41336877,
year = {2025},
author = {Torgersen, EL and Ragnarson, I and Molinas, M},
title = {Decoding Attention through EEG: Paving the Way for BCI Applications in Attention-Related Disorders.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11252840},
pmid = {41336877},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; *Attention Deficit Disorder with Hyperactivity/physiopathology/diagnosis ; *Attention/physiology ; *Brain-Computer Interfaces ; Male ; Female ; Adult ; Machine Learning ; Young Adult ; Brain/physiopathology ; },
abstract = {This study investigates attention-related traits in EEG signals to assess the potential of Electroencephalography (EEG) as an objective diagnostic tool for attention-related disorders such as ADHD, anxiety, and learning disabilities. EEG data were collected from 31 participants, including individuals with ADHD, while they performed a Go/No-Go task designed to evaluate attention and impulsivity. The analysis focused on the spectral characteristics of brain activity, examining the relative power of theta, alpha, and beta frequency bands, along with the theta-to-beta ratio (TBR), to identify distinguishing patterns of attention-related brain activity. Results indicate that the ADHD group exhibited higher theta power and consistently elevated TBR, particularly in the Frontal, Temporal, and Occipital brain regions. Machine learning models, such as K-Nearest Neighbors, effectively classified ADHD and Control groups based on TBR with high accuracy. Additionally, the ADHD group demonstrated faster reaction times but made more errors on the Go/No-Go task, highlighting difficulties with sustained attention. These findings suggest that this approach holds promise for developing objective diagnostic tools for attention-related disorders. While some limitations exist, this study underscores the potential of integrating EEG with machine learning to create brain-computer interface (BCI) systems for assessing attention processes.},
}

Humans
*Electroencephalography/methods
*Attention Deficit Disorder with Hyperactivity/physiopathology/diagnosis
*Attention/physiology
*Brain-Computer Interfaces
Male
Female
Adult
Machine Learning
Young Adult
Brain/physiopathology

@article {pmid41336846,
year = {2025},
author = {Pahuja, S and Ivucic, G and Cai, S and De Silva, D and Li, H and Schultz, T},
title = {XAGnet: Cross-Attention Graph Network for Detecting Auditory Attention in Ear-EEG Signals.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-6},
doi = {10.1109/EMBC58623.2025.11252872},
pmid = {41336846},
issn = {2694-0604},
mesh = {*Electroencephalography/methods ; Humans ; *Attention/physiology ; *Signal Processing, Computer-Assisted ; *Ear/physiology ; Algorithms ; *Neural Networks, Computer ; Brain-Computer Interfaces ; },
abstract = {Auditory Attention Detection (AAD) is essential for developing advanced brain-computer interfaces including neuro-steered hearing technologies capable of functioning in complex auditory environments. In this study, we propose XAGnet, a novel method that leverages ear-centered EEG (ear-EEG) data to model both intra-ear and inter-ear neural dependencies for detection of auditory attention to one of the spatial locations. Specifically, Graph Convolutional Networks (GCNs) are applied separately to left and right ear-EEG signals to extract spatial features from each side for intra-ear interactions. A cross-attention mechanism is then introduced to model inter-ear interactions between the left and right ears. The attended features are combined for multi-class classification, with each class representing a speaker or a speaking location. We evaluate our method on a publicly available ear-EEG dataset, involving AAD tasks with four speakers. Experimental results demonstrate that XAGnet outperforms baseline models, highlighting the effectiveness of modeling both intra-ear and inter-ear dependencies in AAD tasks.},
}

*Electroencephalography/methods
Humans
*Attention/physiology
*Signal Processing, Computer-Assisted
*Ear/physiology
Algorithms
*Neural Networks, Computer
Brain-Computer Interfaces

@article {pmid41336840,
year = {2025},
author = {Jahanjoo, A and Wei, Y and Haghi, M and Schorpf, P and TaheriNejad, N},
title = {Hybrid CNN-Transformer Model for Accurate Classification of Human Attention Levels Using Workplace EEG Data.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-6},
doi = {10.1109/EMBC58623.2025.11251604},
pmid = {41336840},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; *Attention/physiology ; *Neural Networks, Computer ; *Workplace ; Signal Processing, Computer-Assisted ; Algorithms ; Brain-Computer Interfaces ; Fourier Analysis ; },
abstract = {Accurately detecting human attention levels is a key challenge in cognitive neuroscience, with broad application value in improving productivity. Although Electroencephalography (EEG) signals are often used to study cognitive states, most studies still rely on data collected in controlled laboratory environments. This paper collects EEG data from employees during their daily work using a commercial single-channel EEG headband, making attention detection closer to real-world applications and increasing its feasibility and promotion value. We propose a new classification method based on a multi-head attention transformer to identify six different attention levels. We first perform a Short-Time Fourier Transform (STFT) on the EEG signal. Subsequently, we constructed a transformer architecture to effectively model long-range dependencies and subtle pattern changes in EEG data using self-attention and stacked encoder layers. Experimental results show that our proposed model achieves 87.37% classification accuracy in the six-level attention classification task, outperforming traditional high-performance methods and demonstrating superior performance compared to existing similar approaches. This achievement not only verifies the potential of the transformer architecture in EEG attention level classification but also provides new possibilities for developing advanced tools in fields such as brain-computer interface (BCI) and cognitive monitoring.},
}

Humans
*Electroencephalography/methods
*Attention/physiology
*Neural Networks, Computer
*Workplace
Signal Processing, Computer-Assisted
Algorithms
Brain-Computer Interfaces
Fourier Analysis

@article {pmid41336806,
year = {2025},
author = {Quiles, V and Polo-Hortiguela, C and Soriano-Segura, P and Ortiz, M and Ianez, E and Azorin, JM},
title = {Design of an Asynchronous BMI with Interpretable Neural Networks for Exoskeleton Control: A Proof of Concept on Data Evolution and Scalability Over One Week.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11251541},
pmid = {41336806},
issn = {2694-0604},
mesh = {Humans ; *Exoskeleton Device ; *Neural Networks, Computer ; *Brain-Computer Interfaces ; Robotics ; Equipment Design ; },
abstract = {This paper presents a concept study of a week-long experimental protocol for controlling a lower-limb exoskeleton via a brain-machine interface. The system employed a neural network adapted from EEGNet that distinguishes motor imagery and resting states in a two-dimensional space under both static and movement conditions. Each day, the model was fine-tuned with that day's training data as well as data from previous days. Daily closed-loop asynchronous evaluations were carried out to assess real-time exoskeleton control performance. The results indicate steady improvements in system accuracy over the week, likely due to the cumulative integration of additional data, which enhanced the neural network-based approach to cognitive state classification in a multi-day setting.Clinical relevance-Incorporating repetitive robotic therapies in which the patient can actively engage in rehabilitation is a core goal of neurorehabilitation. Developing non-invasive brain-machine interfaces that enable an increasingly effective mind-robot connection is of great importance. This work outlines a protocol for creating a brain-machine interface controlled by motor imagery.},
}

Humans
*Exoskeleton Device
*Neural Networks, Computer
*Brain-Computer Interfaces
Robotics
Equipment Design

@article {pmid41336805,
year = {2025},
author = {Yuan, Z and Li, Y and Zhang, H and Liu, X and Li, S and Zhu, Y and Wang, H and Li, J and Wang, H},
title = {Decoding Hybrid EEG-fNIRS Upper Limb Motor Execution with Capsule Dynamic Graph Convolutional Neural Network.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11251587},
pmid = {41336805},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; Spectroscopy, Near-Infrared/methods ; *Neural Networks, Computer ; *Upper Extremity/physiology ; Signal Processing, Computer-Assisted ; Algorithms ; Brain-Computer Interfaces ; Convolutional Neural Networks ; },
abstract = {In this study, we proposed a capsule dynamic graph convolution network (EF-CapsDGCN) for accurate decoding of upper limb motor execution (ME) based on both electroencephalogram (EEG) and functional near-infrared spectroscopy (fNIRS) signals. In EF-CapsDGCN, EEG/fNIRS features are extracted using the same convolutional architecture but different parameter settings. The extracted features from both modalities are then dynamically routed to capsules. Afterwards, the single-modality capsules are concatenated to form EEG-fNIRS multimodal capsules. Each capsule is treated as a graph node, and hidden feature representations are learned through dynamic graph convolution. Finally, after concatenating the original capsules with the learned hidden features, the combined features are passed through multi-head self-attention and then flattened to feed into a fully connected layer for classification. Compared to current state-of-the-art methods such as ANN, DeepConvNet, DNN, and EF-Net, the proposed method demonstrated superior classification performance on the multimodal EEG-fNIRS dataset HYGRIP. Furthermore, our model achieves at least 8% higher classification accuracy in multimodal EEG-fNIRS compared to single modality EEG/fNIRS. These results demonstrate the potential of capsule dynamic graph convolution for the multimodal fusion of EEG and fNIRS. The proposed model is promising for accurately decoding motor execution-based brain computer interfaces with EEG-fNIRS multiple signals. Overall, this study provides an effective solution for multimodal-BCI decoding.Clinical Relevance- This study demonstrates that integrating EEG and fNIRS signals via a capsule dynamic graph convolution network (EF-CapsDGCN) improves upper limb motor execution decoding accuracy by at least 8% compared to single-modality approaches, offering clinicians a more reliable tool for developing brain-computer interface systems to enhance rehabilitation or assistive device control in patients with motor impairments.},
}

Humans
*Electroencephalography/methods
Spectroscopy, Near-Infrared/methods
*Neural Networks, Computer
*Upper Extremity/physiology
Signal Processing, Computer-Assisted
Algorithms
Brain-Computer Interfaces
Convolutional Neural Networks

@article {pmid41336758,
year = {2025},
author = {Cueva, VM and Lotte, F and Bougrain, L and Rimbert, S},
title = {Quantifying Inter- and Intra-Subject Variability of Sensorimotor Desynchronization Induced by Median Nerve Stimulation and Motor Imagery for BCI.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11254356},
pmid = {41336758},
issn = {2694-0604},
mesh = {Humans ; *Brain-Computer Interfaces ; Electroencephalography/methods ; *Median Nerve/physiology ; Male ; *Imagination/physiology ; Adult ; Female ; Electric Stimulation ; *Sensorimotor Cortex/physiology ; },
abstract = {Motor Imagery-based Brain-Computer Interfaces (MI-BCIs) enable users to control external devices by interpreting sensorimotor activity recorded via ElectroEncephaloGraphy (EEG). Median Nerve Stimulation (MNS) has recently emerged as a promising alternative motor task for BCI applications. However, intra- and inter-subject EEG variability remains a major challenge, affecting BCI system reliability. While variability is a well-known issue, its precise sources and impact on different EEG patterns remain unclear, with a lack of formal and quantitative studies of BCI variability. Thus, this study quantifies intra- and inter-subject variability in MNS-induced sensorimotor desynchronization (ERD) and compares it with that of MI. Results show that MI elicits stronger ERD with lower intra-subject variability, suggesting more consistent activation patterns, while inter-subject variability is similar between tasks. Additionally, the variability of classification accuracies based on Riemannian geometry exhibits a similar trend. These findings provide insights into EEG variability and its implications for BCI design. Identifying stable neural patterns could improve MI- and MNS-based BCIs, particularly for applications such as intraoperative awareness monitoring.},
}

Humans
*Brain-Computer Interfaces
Electroencephalography/methods
*Median Nerve/physiology
Male
*Imagination/physiology
Adult
Female
Electric Stimulation
*Sensorimotor Cortex/physiology

@article {pmid41336716,
year = {2025},
author = {Abid, U and Zulfiqar, O and Nazeer, H and Naseer, N and Bo, APL and Khan, H},
title = {fNIRS Based Comparative Study of Classifiers and Feature Selection Techniques for Finger Tapping.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-6},
doi = {10.1109/EMBC58623.2025.11254285},
pmid = {41336716},
issn = {2694-0604},
mesh = {Humans ; Spectroscopy, Near-Infrared/methods ; *Fingers/physiology ; Algorithms ; Support Vector Machine ; Male ; Movement/physiology ; Machine Learning ; Adult ; Female ; },
abstract = {This study seeks to classify five-finger movements using machine learning (ML) algorithms. It also examines how feature optimization methods affect classification performance. The signals of functional near-infrared spectroscopy (fNIRS) were acquired from 20 healthy participants as they performed five different finger movements. The recorded signals are represented by a total of 17 spatial features such as kurtosis, variance, mean, skewness and others. The ML classifiers used in the beginning are Support Vector Machine (SVM) and Extreme Gradient Boosting (XGBoost). Their performance parameters including precision, accuracy, F1-score, recall and processing time are recorded initially for the dataset comprising of all the features. Afterwards, three population-based metaheuristic algorithms Genetic Algorithm (GA), Ant Colony Optimization (ACO) and Particle Swarm Optimization (PSO) are used to determine the top features from the dataset. The same ML classifiers are then applied to the selected feature datasets. Classification performance is significantly improved by optimized features, with GA and PSO outperforming ACO. SVM is beaten by XGBoost, while its accuracy (94.94%) is greatest when adopting GA-optimized features. The study also shows the role played by feature selection in improving the efficiency and accuracy of ML models in neuroimaging applications. It also suggests optimized classification pipelines for brain-computer interface systems.},
}

Humans
Spectroscopy, Near-Infrared/methods
*Fingers/physiology
Algorithms
Support Vector Machine
Male
Movement/physiology
Machine Learning
Adult
Female

@article {pmid41336656,
year = {2025},
author = {Memar, MO and Ziaei, N and Nazari, B and Yousefi, A},
title = {RISE-iEEG: Robust to Inter-Subject Electrodes Implantation Variability iEEG Classifier.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11252788},
pmid = {41336656},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods/instrumentation ; Neural Networks, Computer ; Brain-Computer Interfaces ; *Electrodes, Implanted ; Algorithms ; Signal Processing, Computer-Assisted ; },
abstract = {Intracranial electroencephalography (iEEG) is increasingly used for clinical and brain-computer interface applications due to its high spatial and temporal resolution. However, inter-subject variability in electrode implantation poses a challenge for developing generalized neural decoders. To address this, we introduce a novel decoder model that is robust to inter-subject electrode implantation variability. We call this model RISE-iEEG, which stands for Robust to Inter-Subject Electrode Implantation Variability iEEG Classifier. RISE-iEEG employs a deep neural network structure preceded by a participant-specific projection network. The projection network maps the neural data of individual participants onto a common low-dimensional space, compensating for the implantation variability. In other words, we developed an iEEG decoder model that can be applied across multiple participants' data without requiring the coordinates of electrode for each participant. The performance of RISE-iEEG across multiple datasets, including the Music Reconstruction dataset, and AJILE12 dataset, surpasses that of advanced iEEG decoder models such as HTNet and EEGNet. Our analysis shows that the performance of RISE-iEEG is about 7% higher than that of HTNet and EEGNet in terms of F1 score, with an average F1 score of 0.83, which is the highest result among the evaluation methods defined. Furthermore, Our analysis of the projection network weights reveals that the Superior Temporal and Postcentral lobes are key encoding nodes for the Music Reconstruction and AJILE12 datasets, which aligns with the primary physiological principles governing these regions. This model improves decoding accuracy while maintaining interpretability and generalization.},
}

Humans
*Electroencephalography/methods/instrumentation
Neural Networks, Computer
Brain-Computer Interfaces
*Electrodes, Implanted
Algorithms
Signal Processing, Computer-Assisted

@article {pmid41336644,
year = {2025},
author = {Si, Y and Wang, Z and Zhao, X and Xu, T and Zhou, T and Hu, H},
title = {Sub-Group Partition Strategy for RSVP-based Collaborative Brain-Computer Interfaces.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-5},
doi = {10.1109/EMBC58623.2025.11252828},
pmid = {41336644},
issn = {2694-0604},
mesh = {*Brain-Computer Interfaces ; Humans ; *Electroencephalography/methods ; Algorithms ; Signal Processing, Computer-Assisted ; Reproducibility of Results ; },
abstract = {Collaborative brain-computer interfaces (cBCIs) have demonstrated significant improvements in single-trial electroencephalogram (EEG) classification performance in rapid serial visual presentation (RSVP) tasks. However, it remains unclear how to effectively organize multiple collaborators into sub-groups to optimize system performance. This study introduces a novel sub-group partition strategy for RSVP-based cBCI systems. We first developed intra-individual and inter-individual neural response reproducibility (IINRR) as a metric to estimate subgroup capability in RSVP tasks. Based on this metric, we propose an IINRR-based partition strategy to optimize sub-group composition. Additionally, we introduce a metric called collaborative information processing rate (CIPR) to evaluate overall system performance. Our experiments verified the effectiveness of the proposed strategy on a public RSVP-based cBCI dataset. The results showed that our strategy consistently outperformed random partitioning in both within-session and cross-session scenarios, achieving higher classification performance and system efficiency. These findings suggest the strategy's potential for optimizing group mode in practical RSVP-based cBCI applications.},
}

*Brain-Computer Interfaces
Humans
*Electroencephalography/methods
Algorithms
Signal Processing, Computer-Assisted
Reproducibility of Results

@article {pmid41336643,
year = {2025},
author = {Merino, EC and Sun, Q and Dauwe, I and Carrette, E and Meurs, A and Van Roost, D and Boon, P and Van Hulle, MM},
title = {Medial Wall's Potential in Enhancing Finger Movement Decoding from Electrocorticography (ECoG): A Single-Subject Pilot Study.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-6},
doi = {10.1109/EMBC58623.2025.11252768},
pmid = {41336643},
issn = {2694-0604},
mesh = {Humans ; *Electrocorticography/methods ; *Fingers/physiology ; Pilot Projects ; Movement/physiology ; *Brain-Computer Interfaces ; *Motor Cortex/physiology ; Male ; Adult ; },
abstract = {The next generation of motor brain-computer interfaces (BCIs) will likely benefit from integrating recordings from multiple motor-related brain regions. Among these is the medial wall, yet it remains relatively understudied in the case of finger movement decoding. Using electrocorticographic (ECoG) recordings from a subject implanted both over medial and lateral cortical areas, we first assessed the medial wall's potential for multiclass classification (5 fingers + rest). We achieved a six-class accuracy of 0.46, significantly above chance, with rest trials classified most accurately, followed by thumb movement trials. Several frequency features contributed to decoding, with Local Motor Potentials (LMP) being the most influential one, with distinctive activity already prior to movement onset, and power in the α (8-12 Hz) band aiding in decoding rest trials over finger movement trials. Next, we explored whether combining the best medial wall channel with lateral cortical channels could improve decoding performance. We found a significant accuracy improvement for most lateral channels (from an average of 0.36 to 0.42), except for the channel closest to the finger primary motor region, whose accuracy was already high (0.77). These findings highlight the medial wall's potential for motor decoding and its value as a target region for future motor BCIs, especially for individuals with impaired hand motor areas.},
}

Humans
*Electrocorticography/methods
*Fingers/physiology
Pilot Projects
Movement/physiology
*Brain-Computer Interfaces
*Motor Cortex/physiology
Male
Adult

@article {pmid41336630,
year = {2025},
author = {Wen, Y and An, Y and Chu, M and Chen, S and Lu, X and Guo, H and Yu, J},
title = {Classification of Functional Near-Infrared Spectroscopy Based on Gramian Angular Difference Field and a Temporal-Spatial Feature Fusion Network.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-5},
doi = {10.1109/EMBC58623.2025.11254538},
pmid = {41336630},
issn = {2694-0604},
mesh = {Spectroscopy, Near-Infrared/methods ; Humans ; Algorithms ; Deep Learning ; Brain-Computer Interfaces ; Signal Processing, Computer-Assisted ; Neural Networks, Computer ; },
abstract = {Functional near-infrared spectroscopy (fNIRS) is a non-invasive functional neuroimaging technique widely employed in brain-computer interface (BCI) research and diverse clinical applications. The key challenge in fNIRS applications lies in extracting nonlinear structures and complex patterns from one-dimensional time series data. Gramian angular difference field (GADF) transforms one-dimensional time series into two-dimensional images, providing effective feature representation for subsequent signal classification. However, most studies have not explored the combined effects of image features and time series features. In this paper, we propose a deep learning model, VisiTempNet, which integrates both time series and GADF image features in a temporal-spatial fusion approach. The model first performs convolution on time series data based on delayed hemodynamic responses to highlight key features. It then separates the feature extraction process into two parallel modules, and normalizes and fuses these features with learnable weights, assigning greater importance to the most relevant information for classification. Experimental results show that our model achieved an accuracy of 76.65±2.43% on the open access fNIRS2MW dataset, outperforming all baseline models. This validates the effectiveness of combining image and time series features and demonstrates the superiority of the proposed model.},
}

Spectroscopy, Near-Infrared/methods
Humans
Algorithms
Deep Learning
Brain-Computer Interfaces
Signal Processing, Computer-Assisted
Neural Networks, Computer

@article {pmid41336626,
year = {2025},
author = {Bao, X and Xu, K and Zhu, J and Huang, H and Li, K and Huang, Q and Li, Y},
title = {Seasickness Alleviation based on a Mindfulness Brain-Computer Interface.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-6},
doi = {10.1109/EMBC58623.2025.11254567},
pmid = {41336626},
issn = {2694-0604},
mesh = {Humans ; *Brain-Computer Interfaces ; Electroencephalography ; Male ; *Mindfulness ; Adult ; *Motion Sickness/therapy/prevention & control/physiopathology ; Female ; Young Adult ; Attention ; },
abstract = {Seasickness is a common condition that negatively affects both the experience of passengers and the operating performance of maritime personnel. Techniques aimed at redirecting attention have been proposed to alleviate motion sickness symptoms; however, their effectiveness has not yet been rigorously verified, especially in maritime environments, which present unique challenges due to the prolonged and severe motion conditions. This research introduces a mindfulness brain-computer interface (BCI) specifically designed to redirect attention and alleviate seasickness. The system employs a single-channel headband to record prefrontal electroencephalography (EEG) signals, which are wirelessly transmitted to computing devices for real-time mindfulness assessments. Participants receive feedback in the form of mindfulness scores and audiovisual cues, facilitating a redirection of attention from physical discomfort. In maritime experiments with 43 participants across three sessions, 81.39% reported the BCI's effectiveness, and a substantial reduction in seasickness severity was observed using the Misery Scale (MISC). Together, our work presents the first wearable and nonpharmacological solution for alleviating seasickness, and opens up a brand-new application domain for BCIs.},
}

Humans
*Brain-Computer Interfaces
Electroencephalography
Male
*Mindfulness
Adult
*Motion Sickness/therapy/prevention & control/physiopathology
Female
Young Adult
Attention

@article {pmid41336622,
year = {2025},
author = {Ahmadi, K and Dong, L and Kok, RL and Findeisen, R},
title = {Gaussian Process-Based Surrogate Models for Optimizing Electrode Configurations in HD-tDCS.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11254512},
pmid = {41336622},
issn = {2694-0604},
mesh = {*Transcranial Direct Current Stimulation/instrumentation/methods ; Humans ; Electrodes ; Normal Distribution ; Brain/physiology ; Computer Simulation ; Algorithms ; },
abstract = {High-definition transcranial direct current stimulation (HD-tDCS) is a promising noninvasive neurostimulation technique used in therapeutic applications and brain-machine interfaces. It delivers direct current via multiple scalp electrodes, generating targeted electrical fields to modulate specific brain areas. In the context of HD-tDCS, optimizing electrode placements is challenging due to the complexity of brain anatomy and the vast number of possible configurations. While simulation models enable model-based optimization, continuous electrode positioning is generally computationally prohibitive. We propose Gaussian Process (GP)-based framework for optimizing HD-tDCS, allowing continuous prediction of electric field distributions. Unlike traditional leadfield-based methods, which restrict electrode placement, our approach expands the search space for greater precision. We employ a Sparse Gaussian Process (SGP) approximation, optimized using Block-Coordinate Descent and Subset of Data techniques, to efficiently handle large datasets. Results demonstrate that the SGP-based model significantly enhanced focality for superficial and mid-brain regions, achieving performance comparable to leadfield-based methods for deep brain targets. Overall, this framework offers enhanced stimulation precision and flexibility, supporting the advancement of tDCS in research and clinical contexts.},
}

*Transcranial Direct Current Stimulation/instrumentation/methods
Humans
Electrodes
Normal Distribution
Brain/physiology
Computer Simulation
Algorithms

@article {pmid41336584,
year = {2025},
author = {Caracci, V and Riccio, A and D'Ippolito, M and Galiotta, V and Quattrociocchi, I and Formisano, R and Cincotti, F and Toppi, J and Mattia, D},
title = {Impact of latency jitter correction on offline P300-based classification: a preliminary study for BCI applications in MCS patients.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-6},
doi = {10.1109/EMBC58623.2025.11253369},
pmid = {41336584},
issn = {2694-0604},
mesh = {Humans ; *Brain-Computer Interfaces ; *Event-Related Potentials, P300 ; Male ; Female ; Adult ; Electroencephalography/methods ; *Persistent Vegetative State/physiopathology ; Algorithms ; Middle Aged ; },
abstract = {Disorders of Consciousness (DoC) are clinical conditions characterized by different levels of arousal and awareness, including coma, Unresponsive Wakefulness Syndrome and Minimally Conscious State (MCS). A Brain-Computer Interface (BCI) employs brain signals to establish a non-muscular outward channel, representing a key frontier in the clinical care of individuals in MCS, with high potential to enhance communication and quality of life. The P300-based BCIs, which use the P300 ERP as a control signal, are the most investigated to emulate communication in MCS. However, a reliable control by MCS patients of these BCIs still remains matter of question. One major challenge could be the across trials variability of P300 characteristics, possibly related to attentional fluctuations in this population. The trial-by-trial instability of the P300 peak latency, known as latency jitter, negatively impacts classification performance, and an approach to mitigating this issue involves template matching algorithms (e.g. the Adaptive Wavelet Filtering, AWF) which detect and realign the P300 latency at the single-trial level. This study investigated the offline classification performance using Stepwise Linear Discriminant Analysis (SWLDA) models trained with progressively larger training sets, to discriminate target from non-target stimuli during an active auditory oddball paradigm. Performance from raw and jitter-corrected data, collected from a control group and a group of patients diagnosed as MCS, were compared. Results highlighted the key role of latency jitter correction in the enhancement of performance and classification speed.Clinical Relevance- The findings suggest that jitter correction could improve real-world applicability of P300-BCI systems for individuals with DoC.},
}

Humans
*Brain-Computer Interfaces
*Event-Related Potentials, P300
Male
Female
Adult
Electroencephalography/methods
*Persistent Vegetative State/physiopathology
Algorithms
Middle Aged

@article {pmid41336583,
year = {2025},
author = {Orlandi, M and Rapa, PM and Baracat, F and Benini, L and Donati, E and Benatti, S},
title = {Neural Strategies for Upper Limb Movements: Motor Unit Control during Dynamic Contractions at Increasing Speeds.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11253409},
pmid = {41336583},
issn = {2694-0604},
mesh = {Humans ; Electromyography ; *Upper Extremity/physiology ; Male ; Movement/physiology ; *Muscle Contraction/physiology ; Adult ; *Motor Neurons/physiology ; Muscle, Skeletal/physiology ; Female ; Young Adult ; },
abstract = {Understanding motor unit (MU) behavior in dynamic movements remains a critical gap in neuro-rehabilitation, prosthetics, and human-machine interfaces (HMI). While machine learning applied to surface electromyography (sEMG) enables movement classification, it provides little insight into neural control, limiting the development of more precise and adaptive assistive technologies. Recent studies have demonstrated that MU activity can be accurately extracted using high-density sEMG decomposition under isometric conditions. However, extracting and tracking MUs during dynamic tasks remains challenging due to signal non-stationarity caused by changes in muscle length. This study investigates MU control in the forearm flexor muscles across different contraction velocities (5°/s, 10°/s, 20°/s) and force levels (15% and 25% of the maximum voluntary contraction [MVC]). We investigate whether increases in movement velocity are primarily achieved through MU recruitment strategies or by adjusting the discharge rates of already-recruited units. Our findings show that MU control in the upper limb follows a velocity-dependent modulation pattern (p-value < 0.05), favoring discharge rate adjustments over additional MUs recruitment at higher speeds. We also validate the feasibility of MU tracking in dynamic conditions, opening new opportunities for neurotechnology applications such as HMI.},
}

Humans
Electromyography
*Upper Extremity/physiology
Male
Movement/physiology
*Muscle Contraction/physiology
Adult
*Motor Neurons/physiology
Muscle, Skeletal/physiology
Female
Young Adult

@article {pmid41336567,
year = {2025},
author = {Roy Chowdhury, M and Ding, Y and Sen, S},
title = {SSL-SE-EEG: A Framework for Robust Learning from Unlabeled EEG Data with Self-Supervised Learning and Squeeze-Excitation Networks.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11253365},
pmid = {41336567},
issn = {2694-0604},
mesh = {*Electroencephalography/methods ; Humans ; Brain-Computer Interfaces ; *Supervised Machine Learning ; *Signal Processing, Computer-Assisted ; Algorithms ; Deep Learning ; Neural Networks, Computer ; },
abstract = {Electroencephalography (EEG) plays a crucial role in brain-computer interfaces (BCIs) and neurological diagnostics, but its real-world deployment faces challenges due to noise artifacts, missing data, and high annotation costs. We introduce SSL-SE-EEG, a framework that integrates Self-Supervised Learning (SSL) with Squeeze-and-Excitation Networks (SE-Nets) to enhance feature extraction, improve noise robustness, and reduce reliance on labeled data. Unlike conventional EEG processing techniques, SSL-SE-EEG transforms EEG signals into structured 2D image representations, suitable for deep learning. Experimental validation on MindBigData, TUH-AB, SEED-IV and BCI-IV datasets demonstrates state-of-the-art accuracy (91% in MindBigData, 85% in TUH-AB), making it well-suited for real-time BCI applications. By enabling low-power, scalable EEG processing, SSL-SE-EEG presents a promising solution for biomedical signal analysis, neural engineering, and next-generation BCIs. The code is available at https://github.com/roycmeghna/SSL_SE_EEG_EMBC25.},
}

*Electroencephalography/methods
Humans
Brain-Computer Interfaces
*Supervised Machine Learning
*Signal Processing, Computer-Assisted
Algorithms
Deep Learning
Neural Networks, Computer

@article {pmid41336566,
year = {2025},
author = {Guttmann-Flury, E and Wei, Y and Zhao, S},
title = {Automatic Blink-Based Bad EEG channels Detection for BCI Applications.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11253420},
pmid = {41336566},
issn = {2694-0604},
mesh = {*Brain-Computer Interfaces ; Humans ; *Electroencephalography/methods ; *Blinking/physiology ; Algorithms ; Signal Processing, Computer-Assisted ; Signal-To-Noise Ratio ; Artifacts ; Adult ; Male ; },
abstract = {In Brain-Computer Interface (BCI) applications, noise presents a persistent challenge, often compromising the quality of EEG signals essential for accurate data interpretation. This paper focuses on optimizing the signal-to-noise ratio (SNR) to improve BCI performance, with channel selection being a key method for achieving this enhancement. The Eye-Bci multimodal dataset is used to address the issue of detecting and eliminating faulty EEG channels caused by non-biological artifacts, such as malfunctioning electrodes and power line interference. The core of this research is the automatic detection of problematic channels through the Adaptive Blink-Correction and DeDrifting (ABCD) algorithm. This method utilizes blink propagation patterns to identify channels affected by artifacts or malfunctions. Additionally, segmented SNR topographies and source localization plots are employed to illustrate the impact of channel removal by comparing Left and Right hand grasp Motor Imagery (MI). Classification accuracy further supports the value of the ABCD algorithm, reaching an average classification accuracy of 93.81% [74.81%; 98.76%] (confidence interval at 95% confidence level) across 31 subjects (63 sessions), significantly surpassing traditional methods such as Independent Component Analysis (ICA) (79.29% [57.41%; 92.89%]) and Artifact Subspace Reconstruction (ASR) (84.05% [62.88%; 95.31%]). These results underscore the critical role of channel selection and the potential of using blink patterns for detecting bad EEG channels, offering valuable insights for improving real-time or offline BCI systems by reducing noise and enhancing signal quality.},
}

*Brain-Computer Interfaces
Humans
*Electroencephalography/methods
*Blinking/physiology
Algorithms
Signal Processing, Computer-Assisted
Signal-To-Noise Ratio
Artifacts
Adult
Male

@article {pmid41336553,
year = {2025},
author = {Sen, O and Khalifa, A and Chatterjee, B},
title = {High-Speed Neural Signal Inferencing for Handwritten Character Recognition on a Portable Hardware Device.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11253375},
pmid = {41336553},
issn = {2694-0604},
mesh = {Humans ; *Brain-Computer Interfaces ; *Handwriting ; *Signal Processing, Computer-Assisted/instrumentation ; Algorithms ; },
abstract = {Brain-computer interfaces (BCIs) hold immense potential in assisting individuals with severe motor and communication disabilities by enabling neural signal-based activity recognition, such as handwriting. This study presents the very first implementation of neural signal inference on a portable hardware device, facilitating efficient handwritten character recognition on resource-constrained platforms. Neural signals from a publicly available dataset are processed into neural spike-event data, facilitating the classification of 31 handwritten characters on an NVIDIA Jetson TX2. To enhance model generalization and mitigate overfitting, random noise injection and time-shifting-based data augmentation techniques are applied. The proposed approach utilizes EfficientNetB0 with neural spikes, and achieves 99.17% test accuracy, significantly outperforming previous model results. During high-speed inference, EfficientNetB0 achieved a Word Error Rate (WER) of 0.96% and a Character Error Rate (CER) of 0.2%, with a character decoding latency of 37.5 milliseconds on the Jetson TX2 while processing 100 sentences used in daily life. These results validate the feasibility of accurate high-speed neural decoding on portable edge hardware, highlighting the impact of lightweight machine learning models in BCI applications.},
}

Humans
*Brain-Computer Interfaces
*Handwriting
*Signal Processing, Computer-Assisted/instrumentation
Algorithms

@article {pmid41336530,
year = {2025},
author = {Li, S and Yang, M and Sun, J and Sun, J and Yu, G and Lin, L and Meng, J and Xu, M},
title = {EEG features and suitable decoding algorithm of RSVP-based brain-computer interface in continuous scenes.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11251802},
pmid = {41336530},
issn = {2694-0604},
mesh = {*Brain-Computer Interfaces ; Humans ; *Electroencephalography/methods ; *Algorithms ; Male ; Signal Processing, Computer-Assisted ; Adult ; Female ; Evoked Potentials ; Young Adult ; },
abstract = {Brain-computer interface (BCI) based on rapid serial visual presentation (RSVP) hold significant value for achieving robust target detection through the integration of human and machine. RSVP in continuous scenes presents video materials and is thus much closer to real-world applications, which greatly exceeds traditional discrete-scene RSVP in terms of practicality. However, the similarities and differences in electroencephalography (EEG) features between continuous and discrete scenes have not yet been clearly clarified. And there is a lack of research on decoding algorithms that are more suitable for continuous scenes, which seriously hinders the development of continuous-scene target detection. To solve these problems, this study designed a comparative experiment based on RSVP paradigm in continuous and discrete scenes. Event-related potential (ERP), event-related spectral perturbation (ERSP), and inter-trial coherence (ITC) were used to investigate EEG features induced by distinct scenes. Further, this study used sliding hierarchical discriminant component analysis (sHDCA), shrinkage discriminative canonical pattern matching (SKDCPM) and attention-based temporal convolutional network (ATCNet) to implement target/non-target trial classification. Consequently, continuous scenes exhibited fewer induced ERP components, a shorter latency of P300, and reduced neural oscillation activities in alpha and beta1 bands over the occipital region within 0~0.2s. As for classification, traditional machine learning algorithms obtained significantly lower accuracy in continuous scenes. While ATCNet achieved the best and same level of accuracy in both scenes, indicating its suitability for decoding continuous-scene RSVP. The results contributed to develop more practical RSVP-BCI target detection systems.},
}

*Brain-Computer Interfaces
Humans
*Electroencephalography/methods
*Algorithms
Male
Signal Processing, Computer-Assisted
Adult
Female
Evoked Potentials
Young Adult

@article {pmid41336489,
year = {2025},
author = {Song, Z and Wu, S and Zhou, T and Wang, Y},
title = {Extracting Preserved Neural Latent Dynamics Across Tasks using Convolutional Transformer-based Variational Autoendecoder.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11251780},
pmid = {41336489},
issn = {2694-0604},
mesh = {Rats ; Animals ; Algorithms ; *Neurons/physiology ; *Neural Networks, Computer ; Brain-Computer Interfaces ; },
abstract = {Understanding how neural systems drive behavior is a fundamental goal in neuroscience. Numerous studies have demonstrated that the activity of large neural populations is often governed by low-dimensional neural dynamics. While much of the current research has focused on extracting informative and interpretable latent dynamics from individual motor tasks, it remains unclear whether these dynamics are preserved across different motor tasks. This question is particularly critical, as prior experience with a related task can facilitate faster learning in a new task. In this paper, we propose a Convolutional Transformer-based Variational Autoencoder (Conformer-VAE) to extract preserved neural latent dynamics across tasks by leveraging the rich spatiotemporal patterns in neural activity. We validate our approach using neural recordings from a rat, which first performed a one-lever pressing task (old task) and subsequently a two-lever discrimination task (new task). By projecting the inferred latent dynamics from both tasks onto a common 2D PCA plane, our results demonstrate that Conformer-VAE effectively captures preserved neural dynamics across tasks, outperforming baseline methods. Moreover, these preserved dynamics enable faster decoder training for the new task by transferring the neural-to-movement mapping learned from the old task. This capability facilitates seamless real-time task switching, offering promising applications for brain-machine interface systems.Clinical Relevance-This work facilitates faster adaptation in brain-machine interfaces by preserving neural dynamics across tasks, offering potential benefits for neuroprosthetics and motor rehabilitation in patients with motor impairments.},
}

Rats
Animals
Algorithms
*Neurons/physiology
*Neural Networks, Computer
Brain-Computer Interfaces

@article {pmid41336487,
year = {2025},
author = {Iacomi, F and Tiberio, P and Tonon, T and Perugini, S and Farabbi, A and Barbieri, R and Mainardi, L},
title = {Validation of a Novel Protocol for Whole-Sentence Imagined Speech Acquisition: Advancing Brain-Computer Interface Applications.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11251720},
pmid = {41336487},
issn = {2694-0604},
mesh = {Humans ; *Brain-Computer Interfaces ; *Speech/physiology ; Male ; Female ; *Imagination/physiology ; Young Adult ; Adult ; Electroencephalography/methods ; },
abstract = {This study aims to validate a novel protocol for whole-sentence imagined speech acquisition, building upon and addressing limitations of a previous single-word acquisition protocol. Eight participants (gender-balanced, mean age 21.3±6 years) were recruited for this study. Participant attention indices, and session variations were evaluated across multiple sessions. The protocol successfully maintains participant engagement while effectively stimulating language imagination processes. The neurophysiological findings, particularly the activation patterns in specific frequency bands and cortical regions, align well with established literature on imagined speech processing. The enhanced delta band activation observed during second sessions, associated with memory mechanisms, provides valuable insight into the cognitive processes involved in repeated imagined speech tasks. These findings contribute to the broader field of Brain Computer Interface (BCI) development and suggest potential applications in clinical settings, particularly for individuals with speech impairments.},
}

Humans
*Brain-Computer Interfaces
*Speech/physiology
Male
Female
*Imagination/physiology
Young Adult
Adult
Electroencephalography/methods

@article {pmid41336480,
year = {2025},
author = {Ramiotis, G and Mania, K},
title = {Enhancing EEG Classification for Motor Imagery Control of a VR Game based on Deep Learning Techniques on Small Datasets.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11251707},
pmid = {41336480},
issn = {2694-0604},
mesh = {*Electroencephalography/methods ; Humans ; *Brain-Computer Interfaces ; *Deep Learning ; Algorithms ; *Virtual Reality ; Neural Networks, Computer ; *Video Games ; Signal Processing, Computer-Assisted ; *Imagination ; Movement ; },
abstract = {Motor imagery-based Brain-Computer Interfaces (BCIs) suffer from limited accuracy when the EEG dataset is recorded from naive BCI users due to noisy components. Neural networks capture more robust representations of EEG features, but require large amount of data which is challenging to collect, due to long motor imagery training sessions. On the other hand, linear- and Riemann-based machine learning algorithms achieve above chance-level accuracy on small scale datasets, but, performance degrades on noisy datasets. To address this issue, we implemented a Wasserstein Generative Adversarial Network (WGAN) for data augmentation to prevent overfitting for the deep classifier, while reaching training convergence faster than existing models. For classification, we developed a Convolutional Neural Network (CNN) to eliminate noisy components caused by BCI illiteracy and extract robust temporal representations of EEG features. To evaluate our system, we designed a VR maze game utilizing the proposed BCI system to translate the EEG signal into movement for a playable character. We achieve increased accuracy, compared to conventional machine learning models, with minimal overfitting, on our own dataset, recorded from 16 naive BCI users.},
}

*Electroencephalography/methods
Humans
*Brain-Computer Interfaces
*Deep Learning
Algorithms
*Virtual Reality
Neural Networks, Computer
*Video Games
Signal Processing, Computer-Assisted
*Imagination
Movement

@article {pmid41336461,
year = {2025},
author = {Soriano-Segura, P and Quiles, V and Ortiz, M and Ianez, E and Azorin, JM},
title = {Effect of Electrode Reduction on the Error-Related Potential Detection During the Start of the Gait.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-5},
doi = {10.1109/EMBC58623.2025.11251757},
pmid = {41336461},
issn = {2694-0604},
mesh = {Humans ; *Gait/physiology ; Electrodes ; *Brain-Computer Interfaces ; Male ; Adult ; },
abstract = {Self-correcting Brain-Machine Interfaces based on Motor Imagery (MI-BMIs) using Error-Related Potentials (ErrP) are a promising approach to improve the accuracy of the system and enhancing their feasibility for the neurorehabilitation of patients with spinal cord injuries (SCI). However, these technologies require extensive preparation time, which shortens the therapy session and causes fatigue in the patient even before starting, potentially reducing the therapy's effectiveness. To address this issue, this study evaluates five electrode configurations to determine the impact of electrode reduction on ErrP detection at the beginning of the gait with a lower-limb exoskeleton. The results indicate that reducing the number of electrodes does not significantly affect detection performance but does reduce false positive rates (FPR). Therefore, these findings support the feasibility of using a reduced electrode configuration of 11 electrodes to enhance BMI usability while maintaining detection reliability.Clinical relevance- The long preparation time required for MI-BMI therapies poses a significant challenge. As a result, patients may begin therapy fatigued or experience rapid exhaustion, limiting their engagement in the rehabilitation process. To address this issue, this study explores electrode reduction for ErrP detection as a strategy to minimize preparation time, enhancing the feasibility of MI-BMIs for clinical applications.},
}

Humans
*Gait/physiology
Electrodes
*Brain-Computer Interfaces
Male
Adult

@article {pmid41336460,
year = {2025},
author = {Wang, X and Lai, YH and Chen, F},
title = {EEG-based Syllable-Level Voice Activity Detection.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11251715},
pmid = {41336460},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; *Brain-Computer Interfaces ; Male ; *Voice/physiology ; Adult ; Female ; *Speech/physiology ; Signal Processing, Computer-Assisted ; Young Adult ; },
abstract = {Speech brain-computer interface (BCI), as an ideal means to achieve direct communication between the brain and the outside world, has become a research area of great interest. This work studied syllable-level voice activity detection (VAD) based on electroencephalogram (EEG) signals to help identify the presence or absence of speech-related EEG activity. We utilized EEG signals from 10 participants performing auditory (listening to stimuli) and speech (pronouncing syllables) tasks to measure brain activity. Speech-Based VAD was employed to label the auditory stimuli and voice recordings, generating corresponding brain activity labels, which were then used to classify resting and active (listening or pronouncing) EEG states, respectively. The experimental results showed that the EEG-based VAD model achieved accuracies of 90.93% and 69.57% for the speech production and auditory speech tasks, respectively. The accuracies were lower in the cross-subject classification, with accuracies of 72.63% and 61.15% for the two tasks. Additionally, the experiment further compared the model's performance under different time window conditions, but no significant correlation was found between window length and classification accuracy. This study provided new insights into the application of EEG based speech decoding, particularly in future self-paced speech BCI applications.},
}

Humans
*Electroencephalography/methods
*Brain-Computer Interfaces
Male
*Voice/physiology
Adult
Female
*Speech/physiology
Signal Processing, Computer-Assisted
Young Adult

@article {pmid41336448,
year = {2025},
author = {Liu, G and Yan, Y and Cai, J and Cheok, AD and Qi Wu, E and Song, A},
title = {A More Rational and Efficient Kalman Filter Design for Motor Brain-Machine Interfaces.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-5},
doi = {10.1109/EMBC58623.2025.11251710},
pmid = {41336448},
issn = {2694-0604},
mesh = {*Brain-Computer Interfaces ; Humans ; Algorithms ; },
abstract = {The Kalman Filter has long been one of the most widely used models in motor brain-machine interface (BMI) research due to its noise handling capabilities and real-time adaptability. However, as a model originally developed for traditional control systems, its underlying assumptions of Markov property and the designs of observation models may not always hold true in the context of BMI applications, potentially leading to oversimplifications. This paper examines the limitations that arise when applying the Kalman Filter to BMI, and proposes the Dilated Kalman Filter, which performs Gaussian multiplication between state transition distribution and observation-mapped state distribution in state space, thereby combining observation noise with BMI-specific observation model noise, and consequently incorporates historical information from both states and observations. The proposed method improves the accuracy of Kalman Filter while significantly enhancing computational efficiency, particularly when processing data from large numbers of neurons.},
}

*Brain-Computer Interfaces
Humans
Algorithms

@article {pmid41336444,
year = {2025},
author = {Lin, L and Lin, J and Pu, Q and Zhou, H and Wang, H and Sun, J and Luo, R and Yu, G and Meng, L and He, F and Meng, J and Xu, M},
title = {Regularization SAME Method can Enhance the Performance of SSVEP-BCI with Very Weak Stimulation.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11251722},
pmid = {41336444},
issn = {2694-0604},
mesh = {*Brain-Computer Interfaces ; Humans ; *Evoked Potentials, Visual/physiology ; Electroencephalography/methods ; Algorithms ; Male ; Signal-To-Noise Ratio ; Photic Stimulation ; Adult ; Signal Processing, Computer-Assisted ; Female ; },
abstract = {The steady-state visual evoked potential-based brain-computer interface (SSVEP-BCI) has gained considerable attention due to its high information transfer rate (ITR) and stable performance. However, the comfort of SSVEP-BCI still needs to be improved, as strong flickering stimuli cause users' visual fatigue. Reducing the pixel density of the stimuli has been demonstrated as an effective method to improve its comfort. However, the signal-to-noise rate (SNR) of the SSVEP signal induced by such very weak stimuli is low, posing challenges for their decoding. Therefore, it is necessary to develop suitable strategy for better decoding the SSVEP induced by very weak stimuli. This study employed the source aliasing matrix estimation (SAME) method to enlarge the dataset and improve decoding accuracy for SSVEP induced by low-pixel density stimuli. Additionally, this study further optimized the SAME with a regularization method to achieve much higher decoding performance. A SSVEP experiment was designed with various pixel densities (100%, 90%, 80%, 70%, 60%, 50%, 40%, 30%, 20%, 10% and 1%) and frequencies (low: 7Hz, 11Hz, and 15Hz; mid-to-high: 23Hz, 31Hz, and 39Hz) to verify our methods. The results indicated SAME significantly improved the classification accuracy compared to traditional method without the SAME, especially under very weak stimulation conditions (pixel densities ≤ 50%), with the maximum increase reaching 8.6%. Besides, regularization SAME further yielded a significant enhancement, achieved maximum improvements of 4.29% compared to SAME. The regularization SAME proposed in this study significantly improves SSVEP decoding performance under low-pixel density stimuli, paving the way for the development of comfortable and effective SSVEP-BCI.},
}

*Brain-Computer Interfaces
Humans
*Evoked Potentials, Visual/physiology
Electroencephalography/methods
Algorithms
Male
Signal-To-Noise Ratio
Photic Stimulation
Adult
Signal Processing, Computer-Assisted
Female

@article {pmid41336422,
year = {2025},
author = {Li, H and Zhang, M and Karkkainen, T and Meng, Z},
title = {Single Trial Classification of per-stimulus EEG between Different Speed Accuracy Tradeoffs Instruction.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11254002},
pmid = {41336422},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; Male ; Female ; *Signal Processing, Computer-Assisted ; Adult ; Deep Learning ; Neural Networks, Computer ; Young Adult ; },
abstract = {The speed-accuracy tradeoff represents a cornerstone concept in cognitive processing, highlighting the inherent trade-off between decision-making speed and accuracy. Patients may have different speed-accuracy strategies during their neurologic consultation due to differences in understanding of instructions or increased diagnostic time. Despite extensive investigations into the neural mechanisms underpinning speed-accuracy trade-off (SAT), the classification of neural data to differentiate between distinct SAT strategies remains largely unexplored. This study bridges this critical gap by implementing a deep learning framework to classify single-trial EEG signals based on participants' instructed response strategies-either prioritizing speed or accuracy and leveraging a dataset from 20 participants engaged in a mirror-image judgment task. The data underwent preprocessing and were subsequently transformed using continuous wavelet transformation to extract time-frequency features. Employing a channel-stacking technique, we organized the EEG data into RGB-like images, which were then input into a RegNet convolutional neural network for classification. Ten-fold cross-validation results demonstrated that the occipital region achieved the highest classification accuracy (85.37%), followed by the parietal (82.97%), frontal (80.46%), and central regions (78.57%). This study not only validates the feasibility of single-trial EEG classification in distinguishing between speed and accuracy strategies but also highlights its potential applications in adaptive brain-computer interfaces and cognitive neuroscience research.Clinical Relevance- This study provides a novel method for real-time identification of cognitive strategies (speed vs. accuracy prioritization) via EEG, offering clinicians a tool to tailor neurofeedback or rehabilitation protocols based on individualized neural signatures.},
}

Humans
*Electroencephalography/methods
Male
Female
*Signal Processing, Computer-Assisted
Adult
Deep Learning
Neural Networks, Computer
Young Adult

@article {pmid41336408,
year = {2025},
author = {Hu, G and Zeng, F and Tang, H and Zhao, Y and Zhang, X},
title = {A Study of Brain-Computer Interface Recognition Performance Crossing Action Observation Paradigms.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11254040},
pmid = {41336408},
issn = {2694-0604},
mesh = {*Brain-Computer Interfaces ; Humans ; Electroencephalography/methods ; Male ; Adult ; Female ; Evoked Potentials, Visual/physiology ; Movement/physiology ; },
abstract = {Action observation-based brain-computer interface (AO-BCI) could induce visual motor imagery through biological motion while relying on its movement frequency to stimulate steady-state visual evoked potential. This hybrid BCI with dual-brain-region activation offers significant potential for stroke rehabilitation. Since varying AO paradigms are employed in the rehabilitation of different limb movements, a limited training dataset can compromise recognition performance. Thus, this study tried to investigate the BCI performance crossing different AO paradigms for the first time. Three AO paradigms, each containing four actions, were designed to establish an online BCI system. Task discriminant component analysis was utilized to analyze the online and offline EEG data. Three training schemes were developed to construct spatial filters including target session (TS) data, source session (SS) data, and a combination of both. Results indicated that the paradigm content significantly affected the recognition performance (F=7.65, p=0.0039). The recognition accuracies of the four actions for each AO paradigm were 71.86%, 89.71%, and 82.71%, respectively. Among the three training schemes, the combined TS and SS data approach notably enhanced recognition accuracy for the AO paradigm with poor performance using TS data alone (p=0.0319). This study demonstrated that EEG data from existing AO paradigms can be used to construct training sets for new paradigms. And combining a small amount of data from the new paradigm could improve the recognition performance. Future research should focus on developing data calibration methods specific to cross-AO paradigms to further enhance recognition accuracy. This work will provide valuable insights for advancing AO-BCI applications in rehabilitation.},
}

*Brain-Computer Interfaces
Humans
Electroencephalography/methods
Male
Adult
Female
Evoked Potentials, Visual/physiology
Movement/physiology

@article {pmid41336362,
year = {2025},
author = {McDorman, RA and Raj Thapa, B and Kim, J and Bae, J},
title = {Transfer Learning in EEG-based Reinforcement Learning Brain Machine Interfaces via Q-learning Kernel Temporal Differences.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11253000},
pmid = {41336362},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; *Brain-Computer Interfaces ; *Machine Learning ; Male ; Adult ; Algorithms ; Female ; Young Adult ; Movement ; },
abstract = {Reinforcement learning based brain machine interfaces (RLBMIs) is an emerging technology with many possible real-time applications. Transfer learning (TL) has proved beneficial as it can improve performance of machine learning algorithms by reusing learned knowledge from similar tasks. However, its application in BMIs has mainly focused on supervised learning approaches. In this study, we investigate the effect of TL in RLBMIs to decode freewill movement related intentions using multichannel scalp electroencephalogram (EEG). We applied TL strategies to Q-learning Kernel Temporal Difference (Q-KTD), which is an algorithm to estimate the action value function, Q, by a nonlinear function approximator using kernel methods. A publicly available EEG dataset recorded while healthy adult participants conduct a key pressing task was used to decode premovement (before movement onset) and movement intention (after movement onset). Differently from most cue-based tasks, participants had freewill to choose the key being pressed, providing unique neural dynamics for decoding. TL was applied between and within subjects to decode the movement related intentions. Significant increase on success rates (p < 0.01) were observed in 96% cases. The success rate increases in each case ranged from 1.39 to 10.69%. These results support the use of TL as an effective way to improve the efficiency of RL-based neural decoder's learning.Clinical Relevance- The improved performance of the neural decoder using transfer learning provides efficient modeling strategy of RLBMIs that can assist patients with neurological disorders.},
}

Humans
*Electroencephalography/methods
*Brain-Computer Interfaces
*Machine Learning
Male
Adult
Algorithms
Female
Young Adult
Movement

@article {pmid41336341,
year = {2025},
author = {Germano, D and Ronca, V and Capotorto, R and Di Flumeri, G and Borghini, G and Giorgi, A and Babiloni, F and Arico, P},
title = {Towards the Correction of Covariate Shift in EEG-Based Passive Brain-Computer Interfaces for Out-of-Lab Applications.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11252974},
pmid = {41336341},
issn = {2694-0604},
mesh = {*Brain-Computer Interfaces ; *Electroencephalography/methods ; Humans ; Signal Processing, Computer-Assisted ; Algorithms ; },
abstract = {The increasing adoption of wearable EEG technology is enabling the development of passive Brain-Computer Interface (pBCI) systems for real-world applications, in the near future, such as Industry 5.0. However, one major challenge in classifying electroencephalographic (EEG) signals in these settings is covariate shift, which occurs when the distribution of the data changes between training and testing sessions due to variations in EEG headset positioning. This study investigates the effectiveness of a linear transformation approach to mitigate the negative effect of covariate shift. Simulations were conducted by using different shift conditions (i.e. deviation of the headset position from the original one), to evaluate (i) the performance of the transformation function used for mitigating the covariate shift occurrence and (ii) the importance that the change of reference and/or channels has on the classification performance. Results show that normalizing covariate shift-affected data (i.e., target) using shift-free data as a template (i.e., source) helps mitigate the negative impact of covariate shift, leading to improved classification performanceThe accuracy loss drops from 14% to 6% in the worst configuration and from 5% to 4% in the best configuration. This improvement is more pronounced when the shift is larger, i.e., when both the reference and channels change between the control dataset and the test dataset. These findings have significant implications for the development of robust and reliable pBCI models for out-of-the-lab contexts.},
}

*Brain-Computer Interfaces
*Electroencephalography/methods
Humans
Signal Processing, Computer-Assisted
Algorithms

@article {pmid41336339,
year = {2025},
author = {Hu, C and Liu, Q and Luo, J and Lu, Y and Jiang, N and Li, G and Huai, Y and Li, Y},
title = {Can ICA-Based Artifact Removal Affect Deep Learning Decoding Accuracy? Yes!.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11253033},
pmid = {41336339},
issn = {2694-0604},
mesh = {Humans ; *Deep Learning ; *Artifacts ; *Electroencephalography/methods ; Brain-Computer Interfaces ; *Signal Processing, Computer-Assisted ; Male ; Female ; Adult ; Stroke/physiopathology ; },
abstract = {Regarding brain-computer interfaces (BCIs), the effectiveness of Independent Component Analysis (ICA) for artifact removal in traditional machine learning-based EEG decoding has been widely implemented. However, its utility in deep learning-based EEG decoding remains understudied. This paper investigated the impact of ICA-based artifact removal on the accuracy of deep learning models for decoding motor imagery and motor execution from EEG signals in short time windows. We employed an ICA-based artifact removal approach named ERASE for automatic artifact removal and evaluated the performance of three decoding approaches: CNN, LSTM, and CEBRA. Compared to before artifact removal, The F1-score improved by averages of 27.90% (CNN), 22.06% (LSTM), and 28.38% (CEBRA) after artifacts removal for motor execution tasks in healthy subjects. For motor imagery tasks in stroke patients,The F1-score improved by averages of 18.90% (CNN), 21.04% (LSTM), and 25.84% (CEBRA). Topographic maps and manifold visualizations further confirmed that ICA enhances the spatial specificity and interpretability of neural signals. These findings suggest that ICA-based artifact removal is a valuable preprocessing step for deep learning-based EEG decoding, particularly in scenarios with significant artifact contamination, offering potential benefits for clinical applications such as stroke rehabilitation.},
}

Humans
*Deep Learning
*Artifacts
*Electroencephalography/methods
Brain-Computer Interfaces
*Signal Processing, Computer-Assisted
Male
Female
Adult
Stroke/physiopathology

@article {pmid41336297,
year = {2025},
author = {Ding, Y and Lee, JH and Zhang, S and Luo, T and Guan, C},
title = {Decoding Human Attentive States from Spatial-temporal EEG Patches Using Transformers.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-5},
doi = {10.1109/EMBC58623.2025.11254148},
pmid = {41336297},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; *Attention/physiology ; Brain-Computer Interfaces ; *Signal Processing, Computer-Assisted ; Deep Learning ; Algorithms ; Neural Networks, Computer ; ROC Curve ; },
abstract = {Learning the spatial topology of electroencephalogram (EEG) channels and their temporal dynamics is crucial for decoding attention states. This paper introduces EEG-PatchFormer, a transformer-based deep learning framework designed specifically for EEG attention classification in Brain-Computer Interface (BCI) applications. By integrating a Temporal CNN for frequency-based EEG feature extraction, a pointwise CNN for feature enhancement, and Spatial and Temporal Patching modules for organizing features into spatial-temporal patches, EEG-PatchFormer jointly learns spatial-temporal information from EEG data. Leveraging the global learning capabilities of the self-attention mechanism, it captures essential features across brain regions over time, thereby enhancing EEG data decoding performance. Demonstrating superior performance, EEG-PatchFormer surpasses existing benchmarks in accuracy, area under the ROC curve (AUC), and macro-F1 score on a public cognitive attention dataset. The code can be found via: https://github.com/yi-ding-cs/EEG-PatchFormer.},
}

Humans
*Electroencephalography/methods
*Attention/physiology
Brain-Computer Interfaces
*Signal Processing, Computer-Assisted
Deep Learning
Algorithms
Neural Networks, Computer
ROC Curve

@article {pmid41336280,
year = {2025},
author = {Hong, J and Rao, P and Wang, W and Chen, S and Najafizadeh, L},
title = {ChatBCI-4-ALS: A High-Performance, LLM-Driven, Intent-Based BCI Communication System for Individuals with ALS.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-6},
doi = {10.1109/EMBC58623.2025.11253329},
pmid = {41336280},
issn = {2694-0604},
mesh = {*Amyotrophic Lateral Sclerosis/physiopathology ; Humans ; *Brain-Computer Interfaces ; Algorithms ; *Communication Devices for People with Disabilities ; Electroencephalography ; Event-Related Potentials, P300 ; Language ; },
abstract = {Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease that leads to significant motor and speech impairments, increasing the need for alternative means of communication to support quality of life. P300 speller brain computer interfaces (BCIs) have shown promise in facilitating non-muscular communication by detecting P300 event-related potentials (ERPs) in response to visual stimuli. However, these systems are generally slow and can not fully address the communication needs of ALS patients, specially, when the primary goal is to convey intent with minimal cognitive load. In this paper, we present ChatBCI-4-ALS, the first intent-based BCI communication system designed for individuals with ALS. ChatBCI-4-ALS leverages large language models (LLMs) and employs a dynamic flash algorithm to enhance typing speed, and enable efficient communication of the user's intent beyond exact lexical matches. Additionally, we introduce new semantic-based quantitative performance metrics to evaluate the effectiveness of intent-based communication. Results from online experiments suggest that ChatBCI-4-ALS achieves record-breaking average spelling speed of 23.87 char/min (with the best case scenario of 42.16 char/min), and a best information transfer rate (ITR) of 128.85 bits/min, marking an advancement in P300 BCI-based communication systems.},
}

*Amyotrophic Lateral Sclerosis/physiopathology
Humans
*Brain-Computer Interfaces
Algorithms
*Communication Devices for People with Disabilities
Electroencephalography
Event-Related Potentials, P300
Language

@article {pmid41336226,
year = {2025},
author = {Kaseler, RL and Andreasen Struijk, LNS},
title = {Harmonic Component Analysis: A novel training-free and asynchronous BCI classification method.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11253522},
pmid = {41336226},
issn = {2694-0604},
mesh = {*Brain-Computer Interfaces ; Humans ; *Electroencephalography/methods ; Adult ; Algorithms ; Male ; Evoked Potentials, Visual/physiology ; Signal Processing, Computer-Assisted ; Female ; },
abstract = {Assistive technologies can provide people with locked-in syndrome independence and improve their quality of life. However, existing brain-computer interfaces (BCI) can be unreliable and require excessive training. Therefore, we investigate the possibility of a training-free BCI that can provide asynchronous and online control of assistive robotic technologies. We propose the harmonic component analysis (HCA), a new training-free classifier for signals with known harmonic characteristics, such as steady-state visually evoked potentials. To validate the HCA, it is compared to the well-known canonical correlation analysis (CCA), using an offline data set of 10 healthy participants who performed cue trials with 16 SSVEP-targets. The HCA achieved better performance than a three-component CCA with up to 74% lower computational cost. For asynchronous control, the HCA achieved a detection accuracy of 85% with an average activation time of 1.6s, against 77% after an average of 1.7s for the CCA. For continuous activation, the HCA achieved a true positive rate of 65% with a false positive rate of 0. 59% from 2 s after cue onset until 5 s after, while the CCA achieved a true positive rate of 59% with a false positive rate of 0. 27%. Thus, the HCA is shown to be a well-suited SSVEP-classifier for systems that require asynchronous classification without the need for a calibration or training-session.},
}

*Brain-Computer Interfaces
Humans
*Electroencephalography/methods
Adult
Algorithms
Male
Evoked Potentials, Visual/physiology
Signal Processing, Computer-Assisted
Female

@article {pmid41336218,
year = {2025},
author = {Ciferri, M and Ferrante, M and Toschi, N},
title = {Optimal Transport and Contrastive Learning for Brain Decoding of Musical Perception.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11253498},
pmid = {41336218},
issn = {2694-0604},
mesh = {*Music ; Humans ; Magnetic Resonance Imaging ; *Brain/physiology/diagnostic imaging ; *Auditory Perception/physiology ; Brain Mapping/methods ; Brain-Computer Interfaces ; Algorithms ; Male ; *Learning ; },
abstract = {Brain decoding aims to reconstruct external stimuli from brain activity, providing insights into the neural representation of cognitive experiences. Music decoding from functional magnetic resonance imaging (fMRI) is particularly challenging due to the complexity of auditory processing and the temporal limitations of fMRI signals. In this study, we introduce a novel decoding framework that improves the alignment between fMRI activity and latent musical representations extracted using a pre-trained multimodal model (CLAP). We propose a dual-loss approach combining Optimal Transport and Contrastive Learning to enhance feature mapping and retrieval accuracy. The first loss ensures structural consistency between brain-predicted and true musical embeddings, while the contrastive loss refines the embedding space by maximizing similarities between corresponding pairs and minimizing non-correspondences. Using fMRI data from five subjects listening to music tracks from the GTZAN dataset, our method achieves improved decoding performance, surpassing traditional regression-based approaches from 22.1% top-1 accuracy to 29.3%. These results highlight the potential of integrating Optimal Transport and Contrastive Learning to improve brain decoding performance, paving the way for extending the approach to different sensory domains and applications in Brain-Computer Interfaces (BCI).Clinical relevance- This study could have clinical implications for understanding auditory processing disorders and developing neurorehabilitation strategies. By elucidating how the brain encodes complex auditory stimuli, this approach may contribute to BCI applications for speech and music perception restoration in individuals with hearing impairments or neurological conditions affecting auditory cognition.},
}

*Music
Humans
Magnetic Resonance Imaging
*Brain/physiology/diagnostic imaging
*Auditory Perception/physiology
Brain Mapping/methods
Brain-Computer Interfaces
Algorithms
Male
*Learning

@article {pmid41336204,
year = {2025},
author = {Padfield, N and Turk, S and Mujahid, K and Camilleri, T and Peng, Y and Camilleri, K},
title = {A Spatio-Spectral Analysis of Decoding Imagined Speech from the Idle State.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11253510},
pmid = {41336204},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; *Speech/physiology ; *Imagination/physiology ; Brain-Computer Interfaces ; Male ; Adult ; Female ; },
abstract = {Studies into speech imagery (SI) classification from electroencephalogram (EEG) data have generally focused on distinguishing imagined words from each other, but accurate discrimination from the idle state, when the user is relaxed, is also necessary for asynchronous brain-computer interfaces (BCIs). In this study, frequency bands and scalp regions most important for distinguishing SI from the idle state were identified and related to underlying neural processes. Power spectral density (PSD) features were extracted from each channel, and a statistical analysis of the features, as well as a classification analysis involving six classifiers, was carried out. The parietal region was identified as the most important scalp region, whilst the delta, theta, and gamma bands were the most important frequency bands. Furthermore, the importance of the alpha band, and of the temporal, frontal-temporal, frontal-central, and parietal regions varied significantly between the SI vs Idle and SI vs SI classification problems, highlighting the importance of including the idle state in SI classification studies.Clinical Relevance-This study identifies frequency bands and scalp regions that are significantly important for the SI vs Idle classification problem, which is important for asynchronous SI BCIs.},
}

Humans
*Electroencephalography/methods
*Speech/physiology
*Imagination/physiology
Brain-Computer Interfaces
Male
Adult
Female

@article {pmid41336201,
year = {2025},
author = {Wimpff, M and Aristimunha, B and Chevallier, S and Yang, B},
title = {Fine-Tuning Strategies for Continual Online EEG Motor Imagery Decoding: Insights from a Large-Scale Longitudinal Study.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11253543},
pmid = {41336201},
issn = {2694-0604},
mesh = {*Electroencephalography/methods ; Humans ; Longitudinal Studies ; Brain-Computer Interfaces ; *Imagination/physiology ; Male ; Signal Processing, Computer-Assisted ; Deep Learning ; Adult ; Female ; Algorithms ; Movement ; },
abstract = {This study investigates continual fine-tuning strategies for deep learning in online longitudinal electroencephalography (EEG) motor imagery (MI) decoding within a causal setting involving a large user group and multiple sessions per participant. We are the first to explore such strategies across a large user group, as longitudinal adaptation is typically studied in the single-subject setting with a single adaptation strategy, which limits the ability to generalize findings. First, we examine the impact of different fine-tuning approaches on decoder performance and stability. Building on this, we integrate online test-time adaptation (OTTA) to adapt the model during deployment, complementing the effects of prior fine-tuning. Our findings demonstrate that fine-tuning that successively builds on prior subject-specific information improves both performance and stability, while OTTA effectively adapts the model to evolving data distributions across consecutive sessions, enabling calibration-free operation. These results offer valuable insights and recommendations for future research in longitudinal online MI decoding and highlight the importance of combining domain adaptation strategies for improving BCI performance in real-world applications.Clinical Relevance-Our investigation enables more stable and efficient long-term motor imagery decoding, which is critical for neurorehabilitation and assistive technologies.},
}

*Electroencephalography/methods
Humans
Longitudinal Studies
Brain-Computer Interfaces
*Imagination/physiology
Male
Signal Processing, Computer-Assisted
Deep Learning
Adult
Female
Algorithms
Movement

@article {pmid41336191,
year = {2025},
author = {Gonzalez-Mitjans, A and Salinas-Medina, A and Toussaint, PJ and Valdes-Sosa, P and Evans, A},
title = {AI-Driven Neurodiagnostics: A Scalable Framework for EEG Anomaly Detection Using a Distributed-Delay Neural Mass Model.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11253534},
pmid = {41336191},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; *Artificial Intelligence ; Epilepsy/diagnosis/physiopathology ; Machine Learning ; Signal Processing, Computer-Assisted ; Brain-Computer Interfaces ; Algorithms ; Neural Networks, Computer ; },
abstract = {The integration of biophysically grounded neural simulations with Artificial Intelligence (AI) has the potential to transform clinical neurodiagnostics by overcoming the inherent challenges of limited pathological EEG datasets. We present a novel AI-driven framework that leverages a Distributed-Delay Neural Mass Model (DD-NMM) to generate synthetic EEG signals replicating both healthy and pathological brain states. Through systematic parameter tuning and domain-specific data augmentation, we enrich the diversity of simulated signals, enabling robust anomaly detection using machine learning techniques. Our approach integrates supervised classification and unsupervised one-class anomaly detection, achieving over 95% accuracy in synthetic tests and over 89% when applied to empirical EEG data from epilepsy patients and healthy volunteers. By providing an engineered solution that bridges computational neuroscience with AI, this framework enhances early seizure detection, adaptive neurofeedback, and brain-computer interface applications. Our results demonstrate that theory-driven simulation, combined with state-of-the-art machine learning, can address critical gaps in medical AI, significantly advancing clinical neuroengineering.Clinical relevance- This study provides a scalable and interpretable AI-driven method for EEG anomaly detection, which can support clinicians in identifying seizure patterns and other neurological disorders with high accuracy. The integration of computational neuroscience with AI-based diagnostics offers a potential pathway for early intervention and personalized neurotherapeutic strategies.},
}

Humans
*Electroencephalography/methods
*Artificial Intelligence
Epilepsy/diagnosis/physiopathology
Machine Learning
Signal Processing, Computer-Assisted
Brain-Computer Interfaces
Algorithms
Neural Networks, Computer

@article {pmid41336140,
year = {2025},
author = {Huang, H and Chen, Z and You, Q and Pan, J and Xiao, J},
title = {Emotion Decoding and Consciousness Evaluation in patients with DOC through EEG Microstate analysis.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11253041},
pmid = {41336140},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; *Emotions/physiology ; *Consciousness Disorders/physiopathology/diagnosis ; Male ; Female ; Adult ; Brain-Computer Interfaces ; *Consciousness ; Middle Aged ; Signal Processing, Computer-Assisted ; },
abstract = {Clinicians commonly employ the Coma Recovery Scale-Revised (CRS-R) as a standard tool for assessing patients with disorders of consciousness (DOC). However, the assessment is easily affected by subjective judgment, and patients with DOC are usually unable to provide adequate behavioral responses. Previous studies have indicated that emotion recognition-based brain-computer interface (BCI) can assist in the assessment of DOC, yet they lack more specific and quantitative indicators. This study is the first to apply electroencephalography (EEG) microstates for emotion recognition in patients with DOC. Specifically, EEG microstates were utilized to capture crucial spatio-temporal features of EEG signals, simplifying the rapidly changing EEG signals into a series of prototype topoplots. In this study, EEG data was recorded from 9 patients with DOC and 11 healthy volunteers. Among healthy participants, our system achieved an average classification accuracy of 94.16%, effectively demonstrating its success in eliciting and recognizing emotions. When applied to patients with DOC, the system yielded an average classification accuracy of 77.94%. The results of this study indicate that EEG microstate dynamics are associated with conscious processing in patients with DOC. However, further validation in a larger patient dataset is required to confirm these preliminary findings.},
}

Humans
*Electroencephalography/methods
*Emotions/physiology
*Consciousness Disorders/physiopathology/diagnosis
Male
Female
Adult
Brain-Computer Interfaces
*Consciousness
Middle Aged
Signal Processing, Computer-Assisted

@article {pmid41336101,
year = {2025},
author = {Wang, X and Wang, L and Ding, Y and Chen, F},
title = {EEG-based Auditory Attention Switch Detection with Multi-scale Gated Attention and Multi-task Learning based Hierarchical Spatiotemporal Networks.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-5},
doi = {10.1109/EMBC58623.2025.11253070},
pmid = {41336101},
issn = {2694-0604},
mesh = {*Electroencephalography/methods ; *Attention/physiology ; Humans ; Signal Processing, Computer-Assisted ; Algorithms ; Signal-To-Noise Ratio ; *Auditory Perception/physiology ; },
abstract = {Auditory attention switch detection (AASD) poses significant challenges for adaptive neurotechnologies, particularly under electroencephalogram (EEG) with low signal-to-noise ratios (SNRs). However, the performance of existing methods is limited due to insufficient feature discriminability and high detection delay. To solve the problem, this paper proposes a Hierarchical Spatiotemporal Network (HSTN) for detecting auditory attention switch from EEG signals. The model employs a hierarchical spatiotemporal encoder to extract spatiotemporal features of EEG signals, integrates short-term transient and long-term dependency information through a multi-scale gated attention mechanism, and synchronously optimizes auditory attention switch detection and auditory attention decoding tasks via a multi-task joint training strategy. Experimental results demonstrate that HSTN significantly outperforms baseline models in both auditory attention switch detection (AASD F1=0.89, accuracy 88.6%) and auditory attention decoding tasks (AAD accuracy 89.3%), with superior model parameter efficiency and inference time. Ablation experiments further validate the critical roles of multi-task learning, gated attention, and multi-scale convolutions. This study provides an efficient solution for auditory attention switch detection in complex auditory scenarios.Clinical Relevance-The study confirms that spatiotemporal feature encoding combined with multi-task joint training significantly enhances performance in EEG attention switch detection, providing a practical technical framework for enabling dynamic sound source enhancement in intelligent hearing aids and auditory brain-computer interface systems.},
}

*Electroencephalography/methods
*Attention/physiology
Humans
Signal Processing, Computer-Assisted
Algorithms
Signal-To-Noise Ratio
*Auditory Perception/physiology

@article {pmid41336083,
year = {2025},
author = {Parashiva, PK and Gangadharan K, S and Vinod, AP},
title = {EEGScaler: A Deep Learning Network to Scale EEG Electrode and Samples for Hand Motor Imagery Speed Decoding.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11251649},
pmid = {41336083},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/instrumentation ; *Deep Learning ; Brain-Computer Interfaces ; Electrodes ; *Hand/physiology ; *Imagination/physiology ; Signal Processing, Computer-Assisted ; Movement ; Algorithms ; Male ; },
abstract = {Motor Imagery (MI)-based Brain-Computer Interface (MI-BCI) systems induce neuroplasticity, promoting rehabilitation in stroke patients. Existing MI-BCI systems decode bilateral MI actions from Electroencephalogram (EEG) data to facilitate motor recovery. However, such systems offer limited degrees of freedom. Decoding kinematics information, such as movement speed can enhance control and provide a more natural interface with the environment. Decoding speed-related information from unilateral MI tasks is challenging due to the significant spatial overlap of neuronal sources and the inherently low spatial resolution of EEG. To address this, we propose EEGScaler, an end-to-end deep learning framework designed to decode slow v/s fast MI tasks by adaptively scaling EEG samples and electrodes with high discriminative value. EEGScaler leverages a Multi-Layer Perceptron (MLP) network to assign scale factors to both samples and electrodes. Spatiotemporal features are subsequently extracted using temporal and depth-wise convolution filters. The model is pre-trained on subject-independent data to learn filter weights, while subject-specific fine-tuning further optimizes the MLP-based scaling mechanism. The EEGScaler model performance is evaluated on 14 healthy subjects' data recorded while performing slow v/s fast unilateral MI tasks. The proposed model achieves an average cross-validated accuracy of 65. 98% for decoding fast v/s slow MI speed tasks, outperforming existing methods by approximately 6%. The subject-specific scaling of samples and electrodes using an end-to-end deep learning model for speed from unilateral MI tasks is novel. By effectively decoding movement speed, EEGScaler enhances the degree of freedom in MI-BCI systems, paving the way for more intuitive and efficient neurorehabilitation applications.Clinical Relevance- This advancement has the potential to improve motor rehabilitation strategies by enabling more precise and adaptive BCI-driven therapy tailored to individual recovery needs.},
}

Humans
*Electroencephalography/instrumentation
*Deep Learning
Brain-Computer Interfaces
Electrodes
*Hand/physiology
*Imagination/physiology
Signal Processing, Computer-Assisted
Movement
Algorithms
Male

@article {pmid41336067,
year = {2025},
author = {Ahmadi, H and Mesin, L},
title = {Decoding Visual Imagination and Perception from EEG via Topomap Sequences.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11251641},
pmid = {41336067},
issn = {2694-0604},
mesh = {*Electroencephalography/methods ; Humans ; *Imagination/physiology ; *Visual Perception/physiology ; Brain-Computer Interfaces ; Signal Processing, Computer-Assisted ; Neural Networks, Computer ; Adult ; Male ; },
abstract = {We propose a Topomap-based EEG decoding framework for distinguishing pictorial Imagination from Perception. By converting each trial's EEG signals into dense sequences of scalp voltage maps at short time intervals, our approach captures crucial spatiotemporal patterns that standard methods may overlook. We then apply a CNN with squeeze-and-excitation (SE) blocks to these Topomap "frames," enabling direct learning of both spatial topographies and rapid temporal fluctuations. Despite using only one trial per subject to simulate a data-scarce scenario, our model achieves 95.1% accuracy under a leave-one-subject-out (LOSO) cross-validation scheme. Results indicate clear neural distinctions between Imagination and Perception states, reflecting focused brain-region engagement during visual recall. In addition to confirming the viability of Topomaps as EEG feature representations, this study underscores their potential generalizability. We anticipate future extensions incorporating other modalities (orthographic, audio) and more advanced deep architectures will further expand the utility and robustness of this approach for brain-computer interface (BCI) applications.Clinical relevance- This framework offers a robust method for accurately distinguishing visual Imagination from Perception, even in data-scarce scenarios. It holds potential for enhancing diagnostic tools in cognitive disorders and refining BCI applications in clinical settings.},
}

*Electroencephalography/methods
Humans
*Imagination/physiology
*Visual Perception/physiology
Brain-Computer Interfaces
Signal Processing, Computer-Assisted
Neural Networks, Computer
Adult
Male

@article {pmid41336065,
year = {2025},
author = {Perley, AS and Coleman, TP},
title = {A Dynamic Mutual Information Measure of Phase Amplitude Coupling.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11251622},
pmid = {41336065},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; Algorithms ; *Signal Processing, Computer-Assisted ; Linear Models ; Sleep/physiology ; },
abstract = {Phase-amplitude coupling (PAC) is a fundamental neural phenomenon in which the phase of a slow oscillation modulates the amplitude of a faster oscillation. PAC has been implicated in various cognitive and clinical conditions, including Parkinson's disease, epilepsy, and depression. Traditional methods for quantifying PAC compute a single summary statistic over an entire time series, limiting their ability to capture dynamic fluctuations. Growing interest in time-varying PAC has led to methods that rely on windowed time-series analysis, but these approaches struggle to track rapid changes in coupling at single-sample resolution. To address this limitation, we propose a novel dynamic mutual information measure of PAC, leveraging a state-space modeling approach based on a Gamma generalized linear model (GLM). By introducing a Gauss-Markov process on the regression weights, our method enables dynamic, interpretable PAC estimation at each time point. We validate our approach using synthetic phase-amplitude coupled signals with time-varying coupling coefficients and demonstrate superior performance in smoothly tracking PAC over time and distinguishing coupled from uncoupled states. Additionally, we apply our technique to sleep EEG data, successfully identifying PAC during sleep spindles, which may serve as a biomarker for neurophysiological conditions such as Alzheimer's disease. Our findings suggest that this dynamic PAC measure is a powerful tool for neuroscientific and clinical research, with potential applications in real-time brain-computer interfaces and neurostimulation protocols.Clinical relevanceThis work demonstrates a new technique for quantifying time-varying electrophysiological coupling. This may allow for understanding transient neural dynamics in disease states and may help more robustly inform electrical stimulation protocols for patients with neurodegenerative disorders.},
}

Humans
*Electroencephalography/methods
Algorithms
*Signal Processing, Computer-Assisted
Linear Models
Sleep/physiology

@article {pmid41335991,
year = {2025},
author = {Li, H and Xu, G and Zhang, S and Xie, J and Han, C and Wu, Q and Zhang, S},
title = {Signal extension with SeU-net for boosting the decoding performance of short-time SSVEP-based brain-computer interfaces.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-6},
doi = {10.1109/EMBC58623.2025.11253264},
pmid = {41335991},
issn = {2694-0604},
mesh = {*Brain-Computer Interfaces ; Humans ; *Evoked Potentials, Visual/physiology ; *Signal Processing, Computer-Assisted ; Electroencephalography/methods ; Algorithms ; },
abstract = {Steady-state visual evoked potential (SSVEP)-based brain-computer interfaces (SSVEP-BCIs) have greatly benefited the lives of patients. However, existing SSVEP recognition methods exhibit poor performance on short SSVEP signals. SSVEP recognition accuracy heavily depends on signal length, which increases as the signal length. From a novel data perspective, this study proposes a signal extension method called SeU-net without requiring calibration data from the target subject to improve the recognition performance of calibration-free methods for short-time SSVEP signals. SeU-net employs LSTM and contrastive learning to enhance feature extraction, converting signals from sample space to feature space, and then back to the sample space to realize signal extension. SeU-net is designed to focus only on signal extension in the temporal domain, without subject-specific feature extraction operations, resulting in strong cross-subject signal extension performance. The extensive experiments demonstrate that SeU-net significantly enhances the decoding performance of calibration-free methods for short-time SSVEP signals. By enabling more accurate decoding with shorter SSVEP signals, SeU-net holds the potential to advance the practical application of high-speed SSVEP-BCIs further.},
}

*Brain-Computer Interfaces
Humans
*Evoked Potentials, Visual/physiology
*Signal Processing, Computer-Assisted
Electroencephalography/methods
Algorithms

@article {pmid41335988,
year = {2025},
author = {Kulwa, F and Sarwatt, DS and Asogbon, MG and Huang, J and Khushaba, RN and Oyemakinde, TT and Li, G and Samuel, OW and Li, H and Li, Y},
title = {A Novel Levant's Differentiator-Based Descriptor for EEG-Based Motor Intent Decoding.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-6},
doi = {10.1109/EMBC58623.2025.11253249},
pmid = {41335988},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; *Brain-Computer Interfaces ; Stroke/physiopathology ; Male ; Signal Processing, Computer-Assisted ; Stroke Rehabilitation ; Algorithms ; Movement ; Female ; },
abstract = {Motor intent (MI)-based brain-computer interfaces (BCIs) have been extensively studied to improve the performance and clinical realization of assistive robots for motor recovery in stroke patients. However, challenges arise in their low decoding performance. This can be attributed to the low spatial resolution and signal-to-noise ratio of electroencephalography (EEG), particularly in accurately deciphering hand movements, which reduces classification performance. Therefore, we have developed a novel feature extraction technique that exploits Levant's differentiators to extract distinct patterns in EEG signals and employs symmetric positive definite matrices (SPD) to effectively leverage the spatial-temporal properties of the EEG signal. Results from nine post-stroke patients and fifteen normal subjects showed an improved decoding accuracy of 99.16±0.64% and 99.30±0.69%, respectively in classifying twenty-four hand motor intents, significantly outperforming existing related methods. Thus, the proposed technique has the potential to greatly enhance the reliability and effectiveness of EEG-based control systems for post-stroke rehabilitation.Clinical Relevance- The outcome of this study can lead to better control of rehabilitation robots and improve the recovery speed of the stroke patients.},
}

Humans
*Electroencephalography/methods
*Brain-Computer Interfaces
Stroke/physiopathology
Male
Signal Processing, Computer-Assisted
Stroke Rehabilitation
Algorithms
Movement
Female

@article {pmid41335965,
year = {2025},
author = {Thomas, A and Cho, Y and Zhao, H and Carlson, T},
title = {MI-CES: An explainable weak labelling approach to example selection for Motor Imagery BCI classification.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11253265},
pmid = {41335965},
issn = {2694-0604},
mesh = {*Brain-Computer Interfaces ; Humans ; *Imagination/physiology ; Electroencephalography/methods ; Algorithms ; Movement/physiology ; },
abstract = {Motor Imagery (MI) Brain Computer Interfaces (BCI) can be used to control assistive devices such as wheelchairs. These systems require a training period to get both the user and the machine to learn and adapt to each other, achieving an acceptable control accuracy. Previous systems have discovered that providing a form of feedback to the user about what the system thinks the user is thinking can increase the effect of training and increase both the control accuracy of the user and the classification accuracy of the BCI system. However, if this feedback is 'incorrect' due to the classifier behind the BCI system having a poor accuracy, this may cause the user to 'incorrectly' adapt to the feedback, providing the system with further poor examples of MI. In this paper, we propose MI-CES, an explainable 'example selection' approach based on the neuro-physiological principle of MI. We found that while using 2 classification techniques, we achieved a statistically significant increase in classification accuracy across 3 datasets that were comprised of both multi-participant and multi-session recordings.},
}

*Brain-Computer Interfaces
Humans
*Imagination/physiology
Electroencephalography/methods
Algorithms
Movement/physiology

@article {pmid41335962,
year = {2025},
author = {Buda, C and Gambosi, B and Toschi, N and Astolfi, L},
title = {A Deep Learning Framework for Multi-Source EEG Localization.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11253252},
pmid = {41335962},
issn = {2694-0604},
mesh = {*Electroencephalography/methods ; *Deep Learning ; Humans ; Algorithms ; Signal Processing, Computer-Assisted ; Neural Networks, Computer ; Brain/physiology ; },
abstract = {Electroencephalography (EEG) provides millisecond-scale resolution of neural activity but struggles to accurately localize multiple concurrent sources, especially in spatially close regions. Classical linear inverse methods, such as MNE, sLORETA, and dSPM, address the ill-posed inverse problem through regularization but often exhibit a "single-source bias", suppressing smaller generators. This paper introduces a deep learning framework designed to robustly identify multiple sources of activity from short EEG segments. Our approach leverages a realistic simulation pipeline that systematically generates EEG recordings from physiologically plausible, distributed current sources. We train a convolutional neural network (ConvNET) on thousands of such simulations, ensuring generalization by using a forward model distinct from that of classical solvers, thereby minimizing the risk of an "inverse crime". We evaluate our ConvNet against nine well-established inverse solvers (MNE, dSPM, sLORETA, eLORETA, LORETA, LAURA, and depth-weighted variants). Benchmarking across multiple synthetic test scenarios demonstrates that our method consistently outperforms traditional solvers, particularly in resolving closely spaced sources, while maintaining or improving accuracy for single-source cases. These results highlight the potential of deep learning to overcome biases in EEG source imaging, offering a more reliable approach for multi-source localization.Clinical relevance- By leveraging deep learning, our approach improves localization accuracy, particularly in closely spaced or deep brain sources, potentially enhancing presurgical planning, brain-computer interfaces, and real-time neurofeed-back applications.},
}

*Electroencephalography/methods
*Deep Learning
Humans
Algorithms
Signal Processing, Computer-Assisted
Neural Networks, Computer
Brain/physiology

@article {pmid41335905,
year = {2025},
author = {Ding, Y and Wang, X and Chen, F},
title = {Enhancing Cross-subject Auditory Attention Detection with Contrastive Learning for EEG Feature Refinement.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11252602},
pmid = {41335905},
issn = {2694-0604},
mesh = {*Electroencephalography/methods ; Humans ; *Attention/physiology ; Brain-Computer Interfaces ; Algorithms ; Signal Processing, Computer-Assisted ; *Machine Learning ; },
abstract = {Electroencephalography (EEG)-based auditory attention detection (AAD) plays a crucial role in recent auditory brain-computer interface applications. However, the performance of AAD models in cross-subject tasks tends to be significantly degraded due to the excessive differences in EEG features across subjects. To address this challenge, we proposed a novel framework, AAD-ContrastNet, that incorporated contrastive learning to refine the temporal features from EEG and reduce the variance of EEG features across subjects. AAD-ContrastNet consists of four main components: (a) an attention-based EEG encoder; (b) a contrastive-learning-based EEG encoder; (c) a feature refinement module; and (d) a classifier. T-SNE visualization results show that combining contrastive learning with cross-attention feature refinement significantly improves the generalization of extracted EEG features. By comparing with SOTA models (i.e., DenseNet-3D and DARNet), we validate the significant effect of AAD-ContrastNet in improving cross-subject decoding accuracy, highlighting its potential in enhancing the robustness and generalization of EEG-based AAD systems.Clinical Relevance- This study demonstrates the potential of contrastive learning in mitigating cross-subject performance degradation, providing a solid foundation for applying generalized auditory brain-computer interface systems.},
}

*Electroencephalography/methods
Humans
*Attention/physiology
Brain-Computer Interfaces
Algorithms
Signal Processing, Computer-Assisted
*Machine Learning

@article {pmid41335889,
year = {2025},
author = {Sun, Y and Zhang, Z and Qi, Q and Li, X and Sun, J and Zhang, K and Zhuang, J and Chen, X and Gao, X},
title = {Beyond Frequency: Leveraging Spatial Features in SSVEP-Based Brain-Computer Interfaces with Visual Animations.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-5},
doi = {10.1109/EMBC58623.2025.11254745},
pmid = {41335889},
issn = {2694-0604},
mesh = {*Brain-Computer Interfaces ; Humans ; *Evoked Potentials, Visual/physiology ; Male ; Female ; Electroencephalography/methods ; Adult ; Young Adult ; },
abstract = {Current research on steady-state visual evoked potential (SSVEP)-based brain-computer interfaces (BCIs) predominantly focuses on utilizing the frequency- and phase-locking characteristics of SSVEP for encoding purposes. In this study, we propose an innovative paradigm wherein SSVEP serves as a marker, integrated with different types of motion animations to identify distinct neural processing pathways associated with these animations. This approach enables the classification of SSVEP-based BCIs without relying on frequency features. We designed six distinct animations corresponding to six behaviors commonly observed in daily life. Each animation was tagged with a uniform 6 Hz stimulus frequency, forming a six-target classification task. Offline testing was conducted with 10 participants. Despite identical frequency components, significant differences in spatial distribution corresponding to the animations were observed, likely due to the behavioral variations in the animations. Classification analysis demonstrated an accuracy of 0.93 within a 6-second window, validating the practical feasibility of this approach. This paradigm offers a novel direction for the advancement of SSVEP-based BCIs, potentially enabling the integration of multi-sensory information.},
}

*Brain-Computer Interfaces
Humans
*Evoked Potentials, Visual/physiology
Male
Female
Electroencephalography/methods
Adult
Young Adult

@article {pmid41335877,
year = {2025},
author = {Lee, SH and Lee, SH and Lee, SW},
title = {EEG-Translator: A Cross-Modality Framework for Subject-Specific EEG and Voice Reconstruction from Imagined Speech.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11254826},
pmid = {41335877},
issn = {2694-0604},
mesh = {Humans ; *Electroencephalography/methods ; *Brain-Computer Interfaces ; *Speech/physiology ; *Voice/physiology ; Signal Processing, Computer-Assisted ; *Imagination/physiology ; Algorithms ; },
abstract = {Non-invasive brain-computer interfaces (BCIs) offer the potential to enable communication for individuals with speech impairments by decoding neural signals through speech-related electroencephalography (EEG) signals. Beyond domain-specific speech EEG decoding, generative approaches that enable cross-domain reconstruction are needed to enhance the overall system performance. Here, we propose a cross-modal EEG translation framework that reconstructs overt speech EEG from imagined speech EEG, for subject-specific speech synthesis. Our approach integrates a diffusion-based model with GAN training to enhance cross-domain EEG reconstruction by preserving both EEG class information and its time-frequency domain properties. In classification tasks, the reconstructed EEG improves class decoding accuracy by 6.2% over the original imagined EEG. Additionally, EEG reconstruction was trained not only on the EEG signal itself but also by incorporating spectrogram-based features, leveraging a fusion of spatial and spectral losses to preserve EEG properties. Beyond EEG reconstruction, category-wise analysis across a multi-speech paradigm dataset reveals variations in decoding performance, offering linguistic insights crucial for the advancement of speech BCI systems. Our findings highlight the potential of diffusion-driven EEG translation in speech BCIs, emphasizing the importance of integrating deep learning methodologies with linguistic insights for improved neural signal reconstruction and interpretation.},
}

Humans
*Electroencephalography/methods
*Brain-Computer Interfaces
*Speech/physiology
*Voice/physiology
Signal Processing, Computer-Assisted
*Imagination/physiology
Algorithms

@article {pmid41335876,
year = {2025},
author = {Wang, A and Zhang, Y and Zhan, G and Zhang, L and Kang, X},
title = {Flexible-Rigid Bonding of Silicon Based Neural Interface for Deep Brain LFP Recording.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11254814},
pmid = {41335876},
issn = {2694-0604},
mesh = {*Silicon/chemistry ; *Brain-Computer Interfaces ; *Brain/physiology ; Electrodes, Implanted ; Humans ; Equipment Design ; Animals ; },
abstract = {Microfabricated silicon neural probes have become the dominant technology in the field of implantable brain-computer interfaces. Mechanical bonding, electroplating, template printing, flip-chip bonding, and welding are prevalent methods for electrode packaging in preparation; however, these techniques often present challenges such as complex processes, elevated temperatures, or increased electrode thickness. We proposed a novel flexible-rigid bonding method for the silicon based neural interface, which markedly reduced the bonding volume compared with the traditional board to board connector. It simplified the assembly process of silicon probes, increased the electrode integration density and facilitated the assembly of the probe and flexible cable. This approach enables the flexible implantation of silicon electrodes in deep brain regions for recording neural signals.},
}

*Silicon/chemistry
*Brain-Computer Interfaces
*Brain/physiology
Electrodes, Implanted
Humans
Equipment Design
Animals

@article {pmid41335863,
year = {2025},
author = {Li, M and Yao, Y and Dong, B and Wang, K and Yu, H and Xu, M and Ming, D},
title = {A Novel Approach to Improve SSVEP-BCI Performance Through Neurofeedback Training.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-6},
doi = {10.1109/EMBC58623.2025.11254803},
pmid = {41335863},
issn = {2694-0604},
mesh = {Humans ; *Brain-Computer Interfaces ; *Neurofeedback/methods ; *Evoked Potentials, Visual/physiology ; Male ; Adult ; Electroencephalography ; Female ; },
abstract = {Brain-Computer interface (BCI), which translates neural activities into commands for external devices, holds significant promise for clinical rehabilitation and assisted movement for individuals with motor disabilities. Among various BCI paradigms, the steady-state visual evoked potential (SSVEP) based BCI garnered considerable attention due to its relatively stable and high-speed communication capabilities. However, a notable portion of the population, referred to as BCI illiteracy, struggles to effectively control BCI systems due to their inability to generate or modulate the neural patterns required for interaction. To address this issue, we proposed a user-centered approach using neurofeedback training (NFT) to improve individual's performance on SSVEP-BCI. As a result, after a five-day training period, significant improvements in SSVEP-BCI performance were only observed in the training group rather than the control group without training. Notably, some subjects initially determined as BCI-illiterate also gained effective control of the BCI system after training. Further analysis revealed that the improvement of SSVEP-BCI performance had a close link with increased power and inter-trial phase coherence of the SSVEP response, indicating that NFT successfully strengthened the user's task-related neural responses. These findings highlight the potential of NFT as a user-centered intervention to improve BCI control performance, offering a promising pathway to address BCI illiteracy and promote the broader application of BCI systems.Clinical Relevance- This study proposes an effective approach to enhancing the controllability of SSVEP-BCI systems, addressing the critical issue of individual control limitations. The developed method demonstrates significant clinical potential for promoting SSVEP-BCI applications, particularly in facilitating communication and device control for patients with severe motor impairments, such as amyotrophic lateral sclerosis (ALS) and locked-in syndrome (LIS).},
}

Humans
*Brain-Computer Interfaces
*Neurofeedback/methods
*Evoked Potentials, Visual/physiology
Male
Adult
Electroencephalography
Female

@article {pmid41335841,
year = {2025},
author = {Ramos, J and Silva, S and Marques, B and Pais-Vieira, M and Stevenson, A and Bras, S},
title = {Empowering Accessibility: Human-Centered Approach to a BCI Home Control for Impaired People.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11253641},
pmid = {41335841},
issn = {2694-0604},
mesh = {Humans ; *Brain-Computer Interfaces ; Electroencephalography ; User-Computer Interface ; *Locked-In Syndrome/physiopathology/rehabilitation ; Self-Help Devices ; Male ; },
abstract = {Brain-Computer Interfaces (BCIs) have shown significant potential for individuals with motor impairments, either by improving physiotherapy treatments or by enabling to perform simple tasks, autonomously. However, much of this progress remains confined to controlled laboratory environments. This study aims to develop a BCI-controlled interface, for real-life scenario, tailored to allow individuals with Locked-In Syndrome (LIS) to interact with their home environment. To ensure system usability, a Human-Centered Design (HCD) approach was adopted prioritizing end-user needs. The interface control system was tested using a BITalino for Electroencephalogram (EEG) acquisition. Preliminary results demonstrated that professionals recognize the system's potential, highlighting the importance of real-time feedback, and design simplicity features to minimize user fatigue and improve control accuracy.Clinical Relevance-This interdisciplinary methodology bridges the gap between assistive technologies and the user needs, promoting autonomy and communication with a BCI-controlled interface for real home interaction.},
}

Humans
*Brain-Computer Interfaces
Electroencephalography
User-Computer Interface
*Locked-In Syndrome/physiopathology/rehabilitation
Self-Help Devices
Male

@article {pmid41335820,
year = {2025},
author = {Luo, R and Zheng, C and Ding, R and Shi, T and Li, D and Xiao, X and Huang, Y and Xu, M and Ming, D},
title = {Boosting Spatial Properties of Single-Flicker SSVEP via Laplacian Electrodes.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-4},
doi = {10.1109/EMBC58623.2025.11253662},
pmid = {41335820},
issn = {2694-0604},
mesh = {Humans ; *Evoked Potentials, Visual/physiology ; *Electroencephalography/instrumentation/methods ; Electrodes ; Male ; Adult ; Female ; Brain-Computer Interfaces ; Photic Stimulation ; },
abstract = {Spatially-encoded steady-state visual evoked potentials (SSVEP) acquired by electroencephalography (EEG) are extensively utilized in brain-computer interface and neuroscience research. However, EEG suffers from low spatial resolution due to volume conduction effects. To tackle this problem, this study developed a bipolar concentric ring electrode (CRE) for collecting high-resolution Laplacian EEG (LEEG), which was validated through a tank simulation experiment and a human experiment. The tank simulation experiment confirmed its high spatial resolution, and the results showed that LEEG acquired by CRE achieved 2.35 times greater spatial attenuation than EEG. Meanwhile, the human experiment designed a single-flicker SSVEP paradigm with stimuli positioned at different visual field orientations. The results revealed that LEEG had lower inter-channel similarity than EEG, with average coefficients of 0.63 for EEG and 0.46 for LEEG (p<0.01). Topographical analysis further demonstrated that CRE sharpened the spatial features of spatially-encoded SSVEPs, and indicated a clear visual hemifield dominance phenomenon. This study effectively enhances the spatial properties of SSVEP and holds promise for advancing high-resolution LEEG.},
}

Humans
*Evoked Potentials, Visual/physiology
*Electroencephalography/instrumentation/methods
Electrodes
Male
Adult
Female
Brain-Computer Interfaces
Photic Stimulation

@article {pmid41335778,
year = {2025},
author = {Nguyen, MTD and Zhu, HY and Burnham, M and Sun, H and Zhu, Q and Nguyen, V and Brown, S and Wu, E and Jin, C and Lin, CT},
title = {Auditory Steady-State Responses and the Effects of Interaural Decoherence and Presence of Vision.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-7},
doi = {10.1109/EMBC58623.2025.11254438},
pmid = {41335778},
issn = {2694-0604},
mesh = {Humans ; Male ; Female ; Adult ; Young Adult ; *Vision, Ocular/physiology ; *Auditory Perception/physiology ; Acoustic Stimulation ; *Evoked Potentials, Auditory/physiology ; Electroencephalography/methods ; },
abstract = {The Auditory Steady-State Response (ASSR) is a periodic neural response used to detect speech and hearing loss, and it is also used as a Brain-Computer Interface paradigm. Our paper identifies two key factors that impact the quality and consistency of the ASSR. First is the interaural decoherence, the timing and intensity of sounds arriving at two ears produced by speakers in reverberant environments. Second is the impact of vision on modulating auditory perception and spatial attention, which could potentially influence the neural synchronisation of the response. To demonstrate this, we conducted an experiment on 26 healthy participants to examine the effects of interaural decoherence, by comparing the frequency responses between speakers and earphones, and the presence of vision, by comparing being blindfolded and non-blindfolded, on the ASSR. This study demonstrates that earphones elicit more consistent and reliable ASSRs compared to speakers, emphasising the detrimental effects of interaural decoherence from speaker-based sound delivery on ASSRs. Furthermore, we found that the response is more biased to one side in the absence of vision compared to the presence of vision. This study highlights the importance of using rooms with anechoic properties or less reverberation when using speakers to ensure the consistency and clarity of the response. Future ASSR paradigms should also consider fixating on a target to elicit less bias in ASSR and more accurate spatial features.},
}

Humans
Male
Female
Adult
Young Adult
*Vision, Ocular/physiology
*Auditory Perception/physiology
Acoustic Stimulation
*Evoked Potentials, Auditory/physiology
Electroencephalography/methods

@article {pmid41335768,
year = {2025},
author = {Morales-Magallon, F and Bojorges-Valdez, E},
title = {Intended and Non-Volitional Knee Joint Movements Elicit Distinct Functional Brain Networks.},
journal = {Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference},
volume = {2025},
number = {},
pages = {1-6},
doi = {10.1109/EMBC58623.2025.11254479},
pmid = {41335768},
issn = {2694-0604},
mesh = {Humans ; *Knee Joint/physiology ; Movement/physiology ; Electroencephalography/methods ; *Brain/physiology ; Male ; Adult ; Brain-Computer Interfaces ; *Nerve Net/physiology ; Female ; },
abstract = {Motor execution induces significant alterations in the dynamics of electroencephalography (EEG) signals, which are crucial for assessing rehabilitation, brain plasticity, and brain-computer interface (BCI) applications. While traditional analyses have primarily focused on power spectral changes, recent advancements incorporate non-linear indices to uncover previously undetected characteristics of brain dynamics.Network analysis provides a powerful framework to examine the structural organization and communication within complex systems composed of interconnected neural units. This study investigates the structural properties functional networks formed during both active and resting states under different knee joint flexion tasks. These movements were performed under three physical demand conditions, including an assisted, non-volitional movement.Functional networks were constructed from EEG analysis over 16 electrodes for the μ, β, and γ frequency bands, and key network metrics were estimated, including input and output node degree centrality, clustering coefficient, and betweenness centrality. Results indicate that motor execution leads to a reduction in overall network connectivity while enhancing communication efficiency. Additionally, networks in the γ and μ bands were more involved in voluntary movement, whereas the β band played a predominant role in assisted movements. The spatial distribution of electrodes contributing to these networks differed between voluntary and assisted conditions, suggesting distinct underlying neural mechanisms rather than a simple linear modulation of connectivity.},
}

Humans
*Knee Joint/physiology
Movement/physiology
Electroencephalography/methods
*Brain/physiology
Male
Adult
Brain-Computer Interfaces
*Nerve Net/physiology
Female