@article {pmid41365192,
year = {2025},
author = {Li, Y and Ye, M and He, Q and Yang, B and Luo, P and Yang, X},
title = {Novel dual AMPK/NRF2 activation by leucocyanidin from Hawthorn (Crataegus) for mitochondria repair-Targeted therapy of hepatic steatosis.},
journal = {Phytomedicine : international journal of phytotherapy and phytopharmacology},
volume = {150},
number = {},
pages = {157614},
doi = {10.1016/j.phymed.2025.157614},
pmid = {41365192},
issn = {1618-095X},
abstract = {BACKGROUND AND PURPOSE: Metabolic dysfunction-associated steatotic liver disease (MASLD) represents a global health challenge with limited therapeutic options. This study identified leucocyanidin (Leuc), a bioactive flavonoid from the traditional herb Crataegus pinnatifida (hawthorn), as a novel dual-target therapeutic agent against MASLD.

METHODS AND RESULTS: We evaluated the effects of Leuc on a mouse model induced by a 60% high-fat diet and a cell model induced by free fatty acids (FFA). Compared to the model group, Leuc treatment dose-dependently significantly reduced liver weight, serum levels of TG and TC, hepatic inflammation markers (IL-6 and TNF-α), as well as cellular TG content. Histological and fluorescence analyses revealed a significant reduction in lipid droplet accumulation. Mechanistically, Leuc exerted its protective effects through two major pathways: (1) By activating the NRF2 antioxidant axis, Leuc attenuated oxidative stress-induced mitochondrial dysfunction and restored fatty acid β-oxidation capacity; (2) Through direct allosteric binding to AMPK, Leuc suppressed fatty acid uptake, inhibited lipogenesis, and enhanced mitochondrial fatty acid transport.

CONCLUSION: These coordinated mechanisms reestablished hepatic lipid homeostasis, positioning Leuc as a promising dual-target natural compound for MASLD intervention through simultaneous AMPK/NRF2 activation.},
}

@article {pmid41364937,
year = {2025},
author = {Patrick-Krueger, KM and Pavlidis, I and Contreras-Vidal, JL},
title = {The state of science convergence in implantable brain-computer interface clinical trials.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/ae2a6f},
pmid = {41364937},
issn = {1741-2552},
abstract = {Advances in implantable brain-computer interfaces (iBCI) have rapidly accelerated in the last decade that promises to improve the quality of life of patients with communications, sensory, and motor control disabilities (CSM). In this Perspective, we quantify the extent and nature of scientific convergence across 21 research groups conducting iBCI clinical trials worldwide. Using Medical Subject Headers (MeSH) and Classification of Instructional Programs (CIP) taxonomies, we analyze topical and disciplinary integration within 161 publications from 1998-2023 to assess how deeply team composition aligns with research themes and translational impact. Our findings indicate uneven patterns of convergence, with many teams combining engineering and clinical expertise yet omitting ethical, legal, and social dimensions. This represents what we term short-cut convergence. We propose an operational definition of this phenomenon and identify practical steps for researchers and funders to strengthen full convergence to accelerate iBCI translation and implementation.},
}

@article {pmid41363017,
year = {2025},
author = {Rayson, H and Moreau, Q and Gailhard, S and Szul, MJ and Bonaiuto, JJ},
title = {Beta Burst Waveform Diversity: A Window onto Cortical Computation.},
journal = {The Neuroscientist : a review journal bringing neurobiology, neurology and psychiatry},
volume = {},
number = {},
pages = {10738584251390779},
doi = {10.1177/10738584251390779},
pmid = {41363017},
issn = {1089-4098},
abstract = {Neural activity in the beta band is increasingly recognized to occur not as sustained oscillations but as transient burst-like events. These beta bursts are diverse in shape, timing, and spatial distribution, but their precise functional significance remains unclear. Here, we review emerging evidence on beta burst properties, functional roles, and developmental trajectories and propose a new framework in which beta bursts are not homogeneous events but reflect distinct patterns of synaptic input from different brain regions targeting different cortical layers. We argue that burst waveform shape carries mechanistic and computational significance, offering a window into the dynamic integration of specific combinations of cortical and subcortical signals. This perspective repositions beta bursts as transient computational primitives, rather than generic inhibitory signals or averaged rhythms. We conclude by outlining key open questions and research priorities, including the need for improved detection methods, investigation into developmental and clinical biomarkers, and translational applications in neuromodulation and brain-computer interfaces.},
}

@article {pmid41362972,
year = {2025},
author = {Labor, VV and Mokienko, OA and Cherkasova, AN and Ikonnikova, ES and Lyukmanov, RK and Suponeva, NA},
title = {[Movement image training and brain-computer interface in cognitive rehabilitation].},
journal = {Zhurnal nevrologii i psikhiatrii imeni S.S. Korsakova},
volume = {125},
number = {11},
pages = {27-35},
doi = {10.17116/jnevro202512511127},
pmid = {41362972},
issn = {1997-7298},
mesh = {Humans ; *Brain-Computer Interfaces ; Parkinson Disease/rehabilitation ; Cognition ; Movement ; Stroke Rehabilitation ; Multiple Sclerosis/rehabilitation ; *Neurological Rehabilitation/methods ; Cognitive Training ; },
abstract = {The paper provides an overview of studies on the use of movement image training and brain-computer interfaces (BCIs) for cognitive rehabilitation in patients with neurological diseases. Based on the analysis of studies published from 2004 to 2025, the effectiveness of these methods in recovering cognitive functions in patients with stroke (13 studies), Parkinson's disease (4 studies), and multiple sclerosis (2 studies) was evaluated. Most studies demonstrated a positive effect of movement image training on the cognitive functions of patients with neurological diseases and moderate cognitive deficits. The effectiveness of this approach is comparable to that of specialized cognitive training. In studies using BCI to control movement image training, an improvement in cognitive functions was also reported. Some studies showed a positive correlation between changes in cognitive indicators and the degree of motor recovery. In groups of patients with normal or near-normal baseline MoCA scores, no significant improvement in cognitive function was reported after a training course. The heterogeneity of the analyzed studies makes direct comparison between them difficult. The results of the analysis of published studies indicate the prospect of using the movement image training with BCI control in the cognitive rehabilitation of neurological patients. However, well-designed randomized controlled trials are necessary to study the mechanisms of the ideomotor training effects on cognitive functions and to develop standardized protocols for assessing their effectiveness.},
}

Humans
*Brain-Computer Interfaces
Parkinson Disease/rehabilitation
Cognition
Movement
Stroke Rehabilitation
Multiple Sclerosis/rehabilitation
*Neurological Rehabilitation/methods
Cognitive Training

@article {pmid41362341,
year = {2025},
author = {Simistira Liwicki, F and Saini, R and Chakladar, DD and Rakesh, S and Gupta, V and Liwicki, M and Eriksson, J},
title = {Simultaneous electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) data during an inner speech task.},
journal = {Data in brief},
volume = {63},
number = {},
pages = {112258},
pmid = {41362341},
issn = {2352-3409},
abstract = {Inner speech, or covert speech, refers to the internal generation of language without overt articulation. Decoding inner speech has significant implications for brain-computer interfaces (BCIs), particularly for assistive communication in individuals with speech and motor impairments. To facilitate research in this area, we introduce a publicly available dataset comprising simultaneously recorded electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) data during inner speech production. Data were collected from three healthy, right-handed participants performing an inner speech task. The task involved silent repetition of visually presented words belonging to either a social or numerical category. The experiment consisted of 40 trials per word, with eight unique words and starts with a fixation period of two seconds. Stimuli were displayed for two seconds at the beginning of each session, followed by a 12-second rest period to allow hemodynamic responses to return to baseline. Participants were instructed to remain still and avoid movements to minimize artifacts. EEG was recorded using a 64-channel MR-compatible cap (BrainCap MR, EasyCap GmbH) at a 5 kHz sampling rate. Electrocardiogram (ECG) signals were simultaneously acquired using an additional electrode placed on the trapezius muscle to facilitate cardioballistic artifact correction. Gradient and cardioballistic artifacts were corrected using BrainVision Analyzer software. Functional MRI data were acquired using a 3T scanner with a 48-channel headcoil, and an echo-planar imaging (EPI) sequence optimized for whole-brain coverage. The repetition time (TR) was 2 s. High-resolution anatomical T1-weighted images were also acquired for structural reference. The dataset is publicly available in the OpenNeuro repository. The aim of this dataset is to provide a resource for studying inner speech processing, multimodal neuroimaging, EEG-fMRI fusion techniques, and BCI-driven speech prosthesis development.},
}

@article {pmid41361966,
year = {2025},
author = {Song, Y and An, S and Choi, Y and Shin, R and Song, KH and Doh, J},
title = {Jammed Foamed Microgel-based Bioprinting for Ex Vivo Reconstruction of 3D T Cell-Cancer Cell Interactions.},
journal = {Advanced healthcare materials},
volume = {},
number = {},
pages = {e05696},
doi = {10.1002/adhm.202505696},
pmid = {41361966},
issn = {2192-2659},
support = {//Korea Health Industry Development Institute (KHIDI)/ ; RS-2023-00208359//National Research Foundation of Korea (NRF)/ ; RS-2023-00218543//National Research Foundation of Korea (NRF)/ ; RS-2023-00217061//National Research Foundation of Korea (NRF)/ ; RS-2024-00512240//Korea Health Industry Development Institute/Republic of Korea ; RS-2024-00406325//Korea Health Industry Development Institute/Republic of Korea ; },
abstract = {T cells in solid tumors migrate through the tumor tissues to find cancer cells and eliminate them. Ex vivo reconstruction of T cell-cancer cell interactions is key for the rational design of cancer immunotherapy. Porous 3D structures essential for optimal T cell motility are challenging to fabricate by 3D printing using conventional bioinks: at high ink concentration, rheological properties are suitable for printing, but T cells are trapped in dense polymer networks, and vice versa. To overcome this limitation, a new bioink based on foamed microgels (FMGs) that facilitates T cell motility is devised, without compromising printability in extrusion 3D printing. Norbornene-functionalized gelatin is synthesized, foamed, cross-linked, and ground to generate FMGs. The FMGs exhibited rougher surfaces than non-foamed microgels (NFMGs), and generated finer pores when jammed. T cell motility is significantly higher in JFMGs than in JNFMGs. Using the JFMG, two compartment structures containing T cells in one compartment and cancer cells in the other compartment are printed. T cells rapidly migrated to the cancer cell compartment and killed the cancer cells. This new bioink enables the ex vivo fabrication of various tissues where immune cell migration is critical.},
}

@article {pmid41361599,
year = {2025},
author = {Wilson, GH and Stein, EA and Kamdar, F and Avansino, DT and Pun, TK and Gross, R and Hosman, T and Singer-Clark, T and Kapitonava, A and Hochberg, LR and Simeral, JD and Shenoy, KV and Druckmann, S and Henderson, JM and Willett, FR},
title = {Long-term unsupervised recalibration of cursor-based intracortical brain-computer interfaces using a hidden Markov model.},
journal = {Nature biomedical engineering},
volume = {},
number = {},
pages = {},
pmid = {41361599},
issn = {2157-846X},
support = {U01DC017844//U.S. Department of Health & Human Services | NIH | National Institute on Deafness and Other Communication Disorders (NIDCD)/ ; U01DC017844//U.S. Department of Health & Human Services | NIH | National Institute on Deafness and Other Communication Disorders (NIDCD)/ ; U01-NS098968//U.S. Department of Health & Human Services | NIH | National Institute of Neurological Disorders and Stroke (NINDS)/ ; U01-NS098968//U.S. Department of Health & Human Services | NIH | National Institute of Neurological Disorders and Stroke (NINDS)/ ; R01-EB028171//U.S. Department of Health & Human Services | NIH | National Institute of Biomedical Imaging and Bioengineering (NIBIB)/ ; 542969//Simons Foundation/ ; },
abstract = {Intracortical brain-computer interfaces (iBCIs) require frequent recalibration to maintain robust performance due to changes in neural activity that accumulate over time, which result in periods when users cannot use their device. Here we introduce a hidden Markov model to infer which targets users are moving towards during iBCI use and we retrain the system using these inferred targets, enabling unsupervised adaptation to changing neural activity. Our approach outperforms distribution alignment methods in large-scale, closed-loop simulations over two months, as well as in a closed loop with a human iBCI user over one month. Leveraging an offline dataset spanning five years of iBCI recordings, we show how target inference recalibration methods appear capable of long-term unsupervised recalibration, whereas recently proposed data-distribution-matching approaches appear to accumulate compounding errors over time. We show offline that our approach performs well on freeform datasets of a person using a home computer with an iBCI. Our results demonstrate the use of task structure to bootstrap a noisy decoder into a highly performant one, thereby overcoming one of the major barriers to clinically translating BCIs.},
}

@article {pmid41361479,
year = {2025},
author = {Vermehren, M and Colucci, A and Angerhöfer, C and Peekhaus, N and Kim, WS and Chang, WK and Kim, H and Hömberg, V and Paik, NJ and Soekadar, SR},
title = {The Berlin bimanual test for stroke survivors (BeBiT-S): evaluating exoskeleton-assisted bimanual motor function after stroke.},
journal = {Journal of neuroengineering and rehabilitation},
volume = {},
number = {},
pages = {},
doi = {10.1186/s12984-025-01822-6},
pmid = {41361479},
issn = {1743-0003},
abstract = {BACKGROUND: Brain/neural hand exoskeletons (B/NHEs) can restore motor function after severe stroke, enabling bimanual tasks critical for various activities of daily living. Yet, reliable clinical tests for assessing bimanual function compatible with B/NHEs are lacking. Here, we introduce the Berlin Bimanual Test for Stroke (BeBiT-S), a 10-task assessment focused on everyday bimanual activities, and evaluate its psychometric properties as well as compatibility with assistive technologies such as B/NHEs.

METHODS: BeBiT-S tasks were selected based on their relevance to daily activities, representation of various grasp types, and compatibility with current (neuro-)prosthetic devices. A scoring system was developed to assess key aspects of bimanual function-including reaching, grasping, stabilizing, manipulating, and lifting-based on video recordings of task performance. The BeBiT-S was administered without support of assistive technology (unassisted condition) to 24 stroke survivors (mean age = 56.5 years; 9 female) with upper-limb hemiparesis. We evaluated interrater reliability through the intraclass correlation coefficient (ICC) and construct validity through correlations with the Chedoke Arm and Hand Activity Inventory (CAHAI), and Stroke Impact Scale (SIS). A subgroup of 15 stroke survivors (mean age 50.3 years, 5 female) completed a second session supported by a B/NHE (B/NHE-assisted condition) to assess the BeBiT-S' sensitivity to change related to B/NHE-application.

RESULTS: The BeBiT-S demonstrated high interrater reliability in both the unassisted (ICC = 0.985, P < .001) and B/NHE-assisted (ICC = 0.862, P < .001) conditions. Unassisted BeBiT-S scores correlated with the CAHAI-8 (r(22) = 0.95, P < .001) and the SIS subscales "strength" (r(20) = 0.53, P = .012) and "hand function" (r(20) = 0.50, P = .018), indicating construct validity. BeBiT-S scores improved significantly with B/NHE assistance (Mdn = 60, P < .05), compared to when no assistance was provided (Mdn = 38, P < .05), demonstrating the test's sensitivity to change following the application of a B/NHE.

CONCLUSIONS: The findings support that the BeBiT-S is a reliable and valid tool for evaluating bimanual task performance in stroke survivors and is compatible with the use of assistive technology such as B/NHEs. Trial registration NCT04440709, submitted June 18th, 2020.},
}

@article {pmid41361196,
year = {2025},
author = {Zhao, R and Bai, Y and Zhang, S and Zhu, J and Liu, H and Ni, G},
title = {An open dataset of multidimensional signals based on different speech patterns in pragmatic Mandarin.},
journal = {Scientific data},
volume = {12},
number = {1},
pages = {1934},
pmid = {41361196},
issn = {2052-4463},
mesh = {Humans ; *Speech ; Electroencephalography ; *Language ; Brain-Computer Interfaces ; Electromyography ; China ; Brain/physiology ; },
abstract = {Speech is essential for human communication, but millions of people lose the ability to speak due to conditions such as amyotrophic lateral sclerosis (ALS) or stroke. Assistive technologies like brain-computer interfaces (BCIs), can convert brain signals into speech. However, these technologies still face challenges in decoding accuracy. This issue is especially challenging for tonal languages like Mandarin Chinese. Furthermore, most existing speech datasets are based on Indo-European languages, which hinders our understanding of how tonal information is encoded in the brain. To address this, we introduce a comprehensive open dataset, which includes multimodal signals from 30 subjects using Mandarin Chinese across overt, silent, and imagined speech modes, covering electroencephalogram (EEG), surface electromyogram (sEMG), and speech recordings. This dataset lays a valuable groundwork for exploring the neural encoding of tonal languages, investigating tone-related brain dynamics, and improving assistive communication strategies. It supports cross-linguistic speech processing research and contributes to data-driven neural speech decoding technology innovations.},
}

Humans
*Speech
Electroencephalography
*Language
Brain-Computer Interfaces
Electromyography
China
Brain/physiology

@article {pmid41361138,
year = {2025},
author = {Wu, M and Yang, Y and Zhang, J and Efimov, AI and Li, X and Zhang, K and Wang, Y and Bodkin, KL and Riahi, M and Gu, J and Wang, G and Kim, M and Zeng, L and Liu, J and Yoon, LH and Zhang, H and Freda, SN and Lee, M and Kang, J and Ciatti, JL and Ting, K and Cheng, S and Zhang, X and Sun, H and Zhang, W and Zhang, Y and Banks, A and Good, CH and Cox, JM and Pinto, L and Vázquez-Guardado, A and Huang, Y and Kozorovitskiy, Y and Rogers, JA},
title = {Patterned wireless transcranial optogenetics generates artificial perception.},
journal = {Nature neuroscience},
volume = {},
number = {},
pages = {},
pmid = {41361138},
issn = {1546-1726},
support = {U01NS131406//U.S. Department of Health & Human Services | NIH | National Institute of Neurological Disorders and Stroke (NINDS)/ ; R01NS107539//U.S. Department of Health & Human Services | NIH | National Institute of Neurological Disorders and Stroke (NINDS)/ ; R01MH117111//U.S. Department of Health & Human Services | NIH | National Institute of Mental Health (NIMH)/ ; 2T32MH06756//U.S. Department of Health & Human Services | NIH | National Institute of Mental Health (NIMH)/ ; R00MH120047//U.S. Department of Health & Human Services | NIH | National Institute of Mental Health (NIMH)/ ; SP-2022-19027//Alfred P. Sloan Foundation/ ; 872599SPI//Simons Foundation/ ; },
abstract = {Synthesizing perceivable artificial neural inputs independent of typical sensory channels remains a fundamental challenge in developing next-generation brain-machine interfaces. Establishing a minimally invasive, wirelessly effective and miniaturized platform with long-term stability is crucial for creating research methods and clinically meaningful biointerfaces capable of mediating artificial perceptual feedback. Here we demonstrate a miniaturized, fully implantable transcranial optogenetic neural stimulator designed to generate artificial perceptions by patterning large cortical ensembles wirelessly in real time. Experimentally validated numerical simulations characterized light and heat propagation, whereas neuronal responses were assessed by in vivo electrophysiology and molecular methods. Cue discrimination during operant learning demonstrated the wireless genesis of artificial percepts sensed by mice, where spatial distance across large cortical networks and sequential order-based analyses of discrimination predicted performance. These conceptual and technical advances expand understanding of artificially patterned neural activity and its perception by the brain to guide the evolution of next-generation all-optical brain-machine communication.},
}

@article {pmid41360829,
year = {2025},
author = {Matran-Fernandez, A and Halder, S},
title = {An EEG dataset to study neural correlates of audiovisual long-term memory retrieval.},
journal = {Scientific data},
volume = {12},
number = {1},
pages = {1933},
pmid = {41360829},
issn = {2052-4463},
mesh = {Humans ; *Electroencephalography ; *Memory, Long-Term ; *Mental Recall ; },
abstract = {Memory retrieval is a fundamental cognitive process that plays a critical role in our lives. Studying the neural correlates of this process has significant implications for numerous fields, such as education and health care. Advances in neuroimaging technologies have facilitated the use of neural data, such as electroencephalography (EEG), to decode cognitive states associated with memory tasks. However, most memory research is still conducted using simple stimuli, such as lists of words, and it is unclear how much the discoveries made with such stimuli generalise to more naturalistic scenarios. We introduce a dataset of EEG signals from 27 participants acquired while they watched 10-second long clips of movies (some of which they had previously seen), together with annotations that reflect whether they recognised or remembered the scenes and the time points of recognition. This dataset allows the study of neural correlates of long-term memory recall in a naturalistic task.},
}

Humans
*Electroencephalography
*Memory, Long-Term
*Mental Recall

@article {pmid41360823,
year = {2025},
author = {Ma, X and Jiang, Y and Jiang, N},
title = {3M-CPSEED, An EEG-based Dataset for Chinese Pinyin Production in Overt, Mouthed, and Imagined Speech.},
journal = {Scientific data},
volume = {},
number = {},
pages = {},
doi = {10.1038/s41597-025-06346-1},
pmid = {41360823},
issn = {2052-4463},
abstract = {Speech brain-computer interfaces (BCIs) enable communication with the external world by decoding neural signals. However, language function as a higher-order brain function, the neural mechanisms underlying speech production remain incompletely understood. Currently most existing Chinese EEG datasets use sentences as stimuli, overlooking that Pinyin constitutes the phonetic foundation of Chinese characters, which limits research on decoding individual Chinese character components. Moreover, most datasets employ only one speech production paradigm, preventing exploration of the brain's diverse speech production modes. This study aims to construct the 3M-CPSEED Chinese Pinyin dataset for exploring neural activity during three distinct speech modes (overt speech, silently articulated speech, imagined speech)of syllables from distinct articulatory positions. The dataset comprises EEG recordings from 20 participants completing four experimental blocks within one day, yielding 1,800 validated trials. 3M-CPSEED holds significant implications for speech neurophysiology research, not only facilitating exploration of neural activity differences across pinyin articulations but also enabling robust transfer learning studies for other alphabetic languages.},
}

@article {pmid41360014,
year = {2025},
author = {Xiong, H and Chang, S and Liu, J},
title = {Dual-Channel TRCA-net based on cross-subject positive transfer for SSVEP-BCI.},
journal = {Biomedical physics & engineering express},
volume = {},
number = {},
pages = {},
doi = {10.1088/2057-1976/ae291c},
pmid = {41360014},
issn = {2057-1976},
abstract = {To enhance the decoding accuracy and information transfer rate of steady-state visual evoked potential-based brain-computer interface (SSVEP-BCI) systems and to reduce inter-subject variability for broader SSVEP-BCI applications, a dual-channel TRCA-net (DC-TRCA-net) method is proposed, based on cross-subject positive transfer. The proposed method incorporates an innovative Transfer-Accuracy-based Subject Selection (T-ASS) strategy and a deep learning network integrated with the SSVEP Domain Adaptation Network (SSVEP-DAN) to enhance SSVEP-BCI decoding performance. The T-ASS strategy constructs contribution scores by computing each subject's self-accuracy and transfer accuracy, and enables effective source subject selection while mitigating negative transfer risks. DC-TRCA-net is further developed to improve model generalization through cross-subject data augmentation. The effectiveness of the proposed method is validated on two large-scale public benchmark datasets. Experimental results demonstrate that DC-TRCA-net outperforms existing networks across both datasets, with particularly substantial performance gains observed in complex experimental scenarios.},
}

@article {pmid41360010,
year = {2025},
author = {Jensen, MA and Schalk, G and Ince, NF and Hermes, D and Worrell, GA and Brunner, P and Staff, NP and Miller, KJ},
title = {sEEG-Based brain-computer interfacing in a large adult and pediatric cohort.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/ae2955},
pmid = {41360010},
issn = {1741-2552},
abstract = {OBJECTIVE: Stereoelectroencephalography (sEEG) is a mesoscale intracranial monitoring technique that records from the brain volumetrically with depth electrodes. sEEG is typically used for monitoring of epileptic foci, but can also serve as a useful tool to study distributed brain dynamics. Herein, we detail the implementation of sEEG-based brain-computer interfacing (BCI) across a diverse and large patient cohort.

APPROACH: Across 27 subjects (15 female, 31 total feedback experiments), we identified channels with increases in broadband power during hand, tongue, or foot movements using a simple block-design screening task. Subsequently, broadband power in these channels was coupled to continuous movement of a cursor on a screen during both overt movement and kinesthetic imagery.

MAIN RESULTS: 26 subjects (29 out of 31 feedback conditions) established successful control, defined as more than 80 percent accuracy, during the overt movement BCI task, while only 12 (of the same 31 conditions) achieved control during the motor imagery BCI task. In successful imagery BCI, broadband power in the reinforced control channel(s) in the two target conditions separated into distinct subpopulations. Outside of the control channel(s), we demonstrate that imagery BCI engages unique subnetworks of the motor system compared to cued movement or kinesthetic imagery alone.

SIGNIFICANCE: Pericentral sEEG-based motor BCI utilizing overt movement and kinesthetic imagery is robust across a diverse patient cohort with inconsistent accuracy during imagined movement. Cued movement, kinesthetic imagery, and feedback engage the motor network uniquely, providing the opportunity to understand the network dynamics underlying BCI control and improve future BCIs.},
}

@article {pmid41360009,
year = {2025},
author = {Faisal, M and Sahoo, S and Hazarika, J},
title = {STeCANet: spatio-temporal cross attention network for brain computer interface systems using EEG-fNIRS signals.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/ae2954},
pmid = {41360009},
issn = {1741-2552},
abstract = {Background- Multimodal neuroimaging fusion has shown promise in enhancing brain-computer interface (BCI) performance by capturing complementary neural dynamics. However, most existing fusion frameworks inadequately model the temporal asynchrony and adaptive fusion between EEG and fNIRS, thereby limiting their ability to generalize across sessions and subjects. Objective- This work aims to develop an adaptive fusion framework that effectively aligns and integrates EEG and fNIRS representations to improve cross-session and cross-subject generalization in BCI applications. Approach- To address this, we propose STeCANet, a novel Spatiotemporal Cross-Attention Network that integrates EEG and fNIRS signals through hierarchical attention-based alignment. The model leverages fNIRS-guided spatial attention, EEG-fNIRS temporal alignment, adaptive fusion, and adversarial training to ensure robust cross-modal interaction and spatiotemporal consistency. Main results- Evaluations across three cognitive paradigms, namely motor imagery (MI), mental arithmetic (MA), and word generation (WG), demonstrate that STeCANet significantly outperforms unimodal and recent multimodal baselines under both session-independent and subject-independent settings. Ablation studies confirm the contribution of each sub-module and loss function, including the domain adaptation component, in boosting classification accuracy and robustness. Significance- These results suggest that STeCANet offers a robust and interpretable solution for next-generation BCI applications.},
}

@article {pmid41359836,
year = {2025},
author = {Wang, Y and Liu, F and Shan, Q and Wang, X and Liu, W and Chen, X and Teng, C and Lv, Y and Gu, X and Wang, X and Yu, B},
title = {Functional recovery induced by KCC2-enabled relay pathways in completely injured spinal cords in adult rats.},
journal = {Proceedings of the National Academy of Sciences of the United States of America},
volume = {122},
number = {50},
pages = {e2421823122},
doi = {10.1073/pnas.2421823122},
pmid = {41359836},
issn = {1091-6490},
mesh = {Animals ; *Spinal Cord Injuries/physiopathology/therapy/metabolism ; *Symporters/metabolism/agonists ; K Cl- Cotransporters ; Rats ; *Recovery of Function/drug effects/physiology ; Neural Stem Cells/transplantation/metabolism ; Rats, Sprague-Dawley ; Female ; Nerve Regeneration/drug effects/physiology ; Axons/physiology ; Locomotion/drug effects ; Insulin-Like Growth Factor I/metabolism ; Osteopontin/metabolism ; },
abstract = {Despite tremendous progress in promoting endogenous axon regeneration and engineering relay pathways by cell transplantation, the obtained functional recovery is still limited. We reason that these regenerated connections might not be able to integrate into the functional circuits in injured spinal cord. In this study, we tested whether modulating the neuronal excitability by pharmacological potassium-chloride cotransporter (KCC2) activation could enhance the functional outcomes of these regenerative treatments in a complete spinal cord injury (SCI) in adult rats. We found that while osteopontin/insulin-like growth factor 1 overexpression (to enhance axon regeneration) or neural stem cell (NSC) transplantation (to build a relay) alone failed to restore the interrupted spinal circuitry, the double treatment facilitated the integration of NSCs into the host spinal network, significantly promoting axonal regeneration and synapse formation. Behavioral assessments demonstrated that the addition of CLP290, a KCC2 agonist, to the combined treatment markedly improved hindlimb locomotion, as evidenced by higher Basso, Beattie and Bresnahan (BBB) scores and enhanced joint oscillation in fine locomotion analysis. Consistently, electrophysiological evaluations indicated partial restoration of electrical transmission through the reconstructed spinal network. Our findings highlight the synergistic effects of KCC2-mediated neuronal modulation on promoting functional recovery after complete SCI.},
}

Animals
*Spinal Cord Injuries/physiopathology/therapy/metabolism
*Symporters/metabolism/agonists
K Cl- Cotransporters
Rats
*Recovery of Function/drug effects/physiology
Neural Stem Cells/transplantation/metabolism
Rats, Sprague-Dawley
Female
Nerve Regeneration/drug effects/physiology
Axons/physiology
Locomotion/drug effects
Insulin-Like Growth Factor I/metabolism
Osteopontin/metabolism

@article {pmid41359725,
year = {2025},
author = {Sun, Y and Chahine, D and Wen, Q and Liu, T and Li, X and Yuan, Y and Calamante, F and Lv, J},
title = {Voxel-Level Brain States Prediction Using Swin Transformer.},
journal = {IEEE journal of biomedical and health informatics},
volume = {29},
number = {12},
pages = {8719-8726},
doi = {10.1109/JBHI.2025.3613793},
pmid = {41359725},
issn = {2168-2208},
mesh = {Humans ; *Magnetic Resonance Imaging/methods ; *Brain/diagnostic imaging/physiology ; *Connectome/methods ; Male ; Female ; Adult ; *Image Processing, Computer-Assisted/methods ; *Signal Processing, Computer-Assisted ; Young Adult ; },
abstract = {Understanding brain dynamics is important for neuroscience and mental health. Functional magnetic resonance imaging (fMRI) enables the measurement of neural activities through blood-oxygen-level-dependent (BOLD) signals, which represent brain states. In this study, we aim to predict future human resting brain states with fMRI. Due to the 3D voxel-wise spatial organization and temporal dependencies of the fMRI data, we propose a novel architecture which employs a 4D Shifted Window (Swin) Transformer as encoder to efficiently learn spatio-temporal information and a convolutional decoder to enable brain state prediction at the same spatial and temporal resolution as the input fMRI data. We used 100 unrelated subjects from the Human Connectome Project (HCP) for model training and testing. Our novel model has shown high accuracy when predicting 7.2s resting-state brain activities based on the prior 23.04s fMRI time series. The predicted brain states highly resemble BOLD contrast and dynamics. This work shows promising evidence that the spatiotemporal organization of the human brain can be learned by a Swin Transformer model, at high resolution, which provides a potential for reducing the fMRI scan time and the development of brain-computer interfaces in the future.},
}

Humans
*Magnetic Resonance Imaging/methods
*Brain/diagnostic imaging/physiology
*Connectome/methods
Male
Female
Adult
*Image Processing, Computer-Assisted/methods
*Signal Processing, Computer-Assisted
Young Adult

@article {pmid41358304,
year = {2025},
author = {Hsieh, TH and Song, H and Shallal, C and Levine, DV and Yeon, SH and Qiao, J and Shu, T and Carty, MJ and McCullough, J and Herr, HM},
title = {Continuous neural control of a 2-DOF ankle-foot prosthesis enables dynamic obstacle maneuvers after transtibial amputation.},
journal = {medRxiv : the preprint server for health sciences},
volume = {},
number = {},
pages = {},
doi = {10.1101/2025.11.25.25340897},
pmid = {41358304},
abstract = {UNLABELLED: Bionic reconstruction techniques that employ surgical neuroprosthetic interfaces, biomimetic control systems, and powered mechatronics have enabled versatile and biomimetic legged gait without reliance on intrinsic gait controllers. However, relative emphasis has been placed on the emulation of sagittal plane biomechanics while neglecting to provide control over frontal plane mechanics critical for terrain adaptation. Here, we present a 2-degree-of-freedom (DOF) bionic reconstruction at the transtibial amputation level that enables continuous neural control of both sagittal and frontal ankle and subtalar joint mechanics. To demonstrate its capabilities in a case study design, we integrated a 2-DOF robotic ankle-foot device via surface electromyographic electrodes to an individual provisioned with a surgical neuroprosthetic interface that augments residual muscle afferents. The subject was able to neurally control both degrees of freedom to regain nominal gait mechanics during both level-ground walking and continuous cross-slope navigation. Furthermore, the subject strategically traversed an obstacle course by dynamically hopping between a series of discrete cross-slope blocks, adapting to the slopes, and responding to rapid foot slips. These preliminary findings suggest that bionic reconstruction techniques can restore continuous neural control over multi-DOF prostheses to achieve agile locomotion over complex terrain.

ONE-SENTENCE SUMMARY: A multi-DOF ankle-foot prosthesis under continuous neural control enables agile locomotion over complex terrain.},
}

@article {pmid41358277,
year = {2025},
author = {Baniasad, A and Chao, S and Nguyen, JA and Tian, E and Modongo, C and Minin, VM and Sebastian, JN and Shin, SS},
title = {HIV Remains a Risk Factor for Unfavorable Tuberculosis Treatment Outcomes in the Era of Universal Access to Antiretroviral Therapy in Botswana.},
journal = {medRxiv : the preprint server for health sciences},
volume = {},
number = {},
pages = {},
doi = {10.1101/2025.11.26.25340699},
pmid = {41358277},
abstract = {Botswana implemented its universal "Treat All" antiretroviral therapy (ART) policy in 2016, expanding treatment eligibility to all people living with HIV (PLHIV). HIV has been known to be a leading risk factor for tuberculosis (TB) and poor TB treatment outcomes. The primary goal of this study is to assess whether HIV infection and HIV-associated immunosuppression remain risk factors for unfavorable TB treatment outcomes in the Post-Treat All era. We analyzed 636 TB patients treated in Gaborone (2017-2023), of whom 54.4% were HIV-positive. Unfavorable outcomes (death, failure, or loss to follow-up) occurred in 19.7% of HIV-positive and 8.5% of HIV-negative patients. We used logistic regression to estimate unadjusted and covariate-adjusted associations between TB treatment outcome and HIV status and between TB treatment outcome and CD4+ T-cell count. HIV-positive patients had 2.5-fold higher odds of unfavorable outcomes compared with HIV-negative patients [adjusted OR: 2.51, 95% CI: (1.48, 4.38)], controlling for age, sex, TB history, distance to clinic, substance use, and occupational status. PLHIV with CD4+ T-cell < 200 cells/ µ L was associated with approximately three-fold higher odds of unfavorable outcomes compared with HIV-negative participants [OR: 3.12, 95% CI: (1.65, 5.97)]. The secondary goal was to test whether the HIV effect changed following Treat All implementation. We combined the data from 2017-2023 with a Pre-Treat All cohort (2012-2016, n= 233, HIV prevalence 60.8%) and fit a frequentist logistic regression and Bayesian mixed-effects models with an interaction term that allows treatment era (Pre- vs. Post-Treat All) to modify the effect of HIV on TB treatment outcome. The estimated change in the HIV effect was uncertain [relative OR: 0.41; 95% CI: (0.11, 1.55)]. Combining the two Botswana data sets with 12 Pre- and Post-Treat All studies from neighboring Ethiopia showed that the pooled effect of HIV infection on unfavorable TB outcome has increased in the Post-Treat All period [relative OR: 2.39; 95% BCI: (1.36, 3.34)].},
}

@article {pmid41357676,
year = {2025},
author = {Akhoundi, A and Yan, P and Landbrug, Y and Hays, M and Murmann, B and Chichilnisky, EJ and Muratore, DG},
title = {A Scalable 1024-Channel Ultra-Low-Power Spike Sorting Chip with Event-Driven Detection and Spatial Clustering.},
journal = {IEEE journal of solid-state circuits},
volume = {60},
number = {11},
pages = {3985-4001},
pmid = {41357676},
issn = {0018-9200},
abstract = {This paper presents a 1024-channel ultra-low-power spike sorting chip featuring event-driven spike detection and spatial clustering for large-scale neural recording. To address power and scalability constraints in brain-computer interfaces, the design integrates a compressive ADC with a two-stage spike detector that significantly reduces memory and processing activity. Spatial features derived from high-density microelectrode array (MEA) enhance cluster separability, enabling robust performance even under neural signal distortion or probe drift, particularly when recordings are obtained using planar MEAs. A modified self-organizing map algorithm clusters spikes in the spatial domain with minimal memory access, supporting on-chip training and real-time operation with low latency. Fabricated in 40 nm CMOS, the chip achieves 0.00029 mm[2]/channel area and 74 nW/channel power consumption, with over 1000× data compression. Performance is validated across synthetic and ex vivo datasets containing up to 500 neurons, demonstrating competitive accuracy and robust drift tracking compared to state-of-the-art solutions with much lower data bandwidth, processing, and power demands.},
}

@article {pmid41356598,
year = {2025},
author = {Chen, S and Xie, N and Tang, Y and Ji, Y and He, Z and Wang, Y and Huang, X and Fu, J and Ge, M and Liu, Q and Li, M and Xiao, Q and Xu, Y and Wang, J and Jia, J and Xu, S},
title = {Long-Term Brain-Computer Interface Functional Electrical Stimulation Enhances Neuroplasticity and Functional Recovery in Elderly Stroke: A 4.5-Year Longitudinal Study Integrating Electroencephalography Biomarkers and Clinical Assessments.},
journal = {Research (Washington, D.C.)},
volume = {8},
number = {},
pages = {0984},
pmid = {41356598},
issn = {2639-5274},
abstract = {Stroke-induced motor and cognitive impairments substantially reduce the quality of life in elderly populations, driving the need for rehabilitation strategies that integrate neural plasticity and functional recovery. In this 4.5-year longitudinal study, we evaluated the efficacy of brain-computer interface combined with functional electrical stimulation (BCI-FES) versus FES only and conventional care (control) in 100 stroke survivors (60 to 90 years; 4,172 total screened, with 24 chronic-stage patients [>1 year post-onset] completing long-term follow-up). We integrated clinical metrics (Fugl-Meyer assessment [FMA], modified Barthel index [MBI], and Montreal Cognitive Assessment [MoCA]) with electroencephalography-based neurophysiological profiling to dissect recovery mechanisms. BCI-FES yielded superior and sustained improvements across all domains: motor function (FMA Δ = 4.5 ± 1.2 points, Cohen's d = 1.2) versus FES (Δ = 1.7 ± 0.8, d = 0.4) and control (Δ = 0.9 ± 0.6, d = 0.2), functional independence (MBI Δ = 5.4 ± 1.5, d = 1.1) exceeding FES (Δ = 2.2 ± 1.1, d = 0.4) and control (Δ = 1.3 ± 0.5, d = 0.5), and cognitive function (MoCA Δ = 1.6 ± 0.5, d = 0.8 at 4 months), although cognitive gains declined to near baseline by 4.5 years. Hemorrhagic stroke patients showed exceptional BCI-FES responses, while ischemic patients exhibited higher variability. Neurophysiologically, BCI-FES induced theta (Cz and C4) and alpha (FC3 and CP3) power increases, with theta power at Cz strongly predicting FMA gains (r = 0.68), and enhanced theta/alpha band functional connectivity (clustering coefficient +22%, local efficiency +18%, and small-world index +15%). Predictive modeling identified that an optimal treatment window (3 to 12 months post-onset with 10 to 15 weeks of therapy) maximizes recovery via peak neuroplasticity, and a responder profile (stroke duration <23 months) includes patients with residual plasticity (age <70, baseline MBI >40), predicting 76% of favorable outcomes. These findings establish BCI-FES as a transformative rehabilitation tool, driving dual-phase recovery via early cortical plasticity and sustained network coherence while highlighting the need for age-tailored cognitive maintenance strategies. This work redefines precision stroke care by merging clinical outcomes with mechanistic insights, positioning BCI-FES as the standard of care for diverse stroke subtypes.},
}

@article {pmid41356332,
year = {2025},
author = {Coutray, K and Barbel, W and Groth, Z and LaViola, JJ},
title = {NeuroGaze: a hybrid EEG and eye-tracking brain-computer interface for hands-free interaction in virtual reality.},
journal = {Frontiers in human neuroscience},
volume = {19},
number = {},
pages = {1695446},
pmid = {41356332},
issn = {1662-5161},
abstract = {Brain-Computer Interfaces (BCIs) have traditionally been studied in clinical and laboratory contexts, but the rise of consumer-grade devices now allows exploration of their use in daily activities. Virtual reality (VR) provides a particularly relevant domain, where existing input methods often force trade-offs between speed, accuracy, and physical effort. This study introduces NeuroGaze, a hybrid interface combining electroencephalography (EEG) with eye tracking to enable hands-free interaction in immersive VR. Twenty participants completed a 360° cube-selection task using three different input methods: VR controllers, gaze combined with a pinch gesture, and NeuroGaze. Performance was measured by task completion time and error rate, while workload was evaluated using the NASA Task Load Index (NASA-TLX). NeuroGaze successfully supported target selection with off-the-shelf hardware, producing fewer errors than the alternative methods but requiring longer completion times, reflecting a classic speed-accuracy tradeoff. Workload analysis indicated reduced physical demand for NeuroGaze compared to controllers, though overall ratings and user preferences were mixed. While the differing confirmation pipelines limit direct comparison of throughput metrics, NeuroGaze is positioned as a feasibility study illustrating trade-offs between speed, accuracy, and accessibility. It highlights the potential of consumer-grade BCIs for long-duration use and emphasizes the need for improved EEG signal processing and adaptive multimodal integration to enhance future performance.},
}

@article {pmid41356065,
year = {2025},
author = {Nair, K and Cecotti, H},
title = {Deep Learning Architectures for Code-Modulated Visual Evoked Potentials Detection.},
journal = {ArXiv},
volume = {},
number = {},
pages = {},
pmid = {41356065},
issn = {2331-8422},
abstract = {Non-invasive Brain-Computer Interfaces (BCIs) based on Code-Modulated Visual Evoked Potentials (C-VEPs) require highly robust decoding methods to address temporal variability and session-dependent noise in EEG signals. This study proposes and evaluates several deep learning architectures, including convolutional neural networks (CNNs) for 63-bit m-sequence reconstruction and classification, and Siamese networks for similarity-based decoding, alongside canonical correlation analysis (CCA) baselines. EEG data were recorded from 13 healthy adults under single-target flicker stimulation. The proposed deep models significantly outperformed traditional approaches, with distance-based decoding using Earth Mover's Distance (EMD) and constrained EMD showing greater robustness to latency variations than Euclidean and Mahalanobis metrics. Temporal data augmentation with small shifts further improved generalization across sessions. Among all models, the multi-class Siamese network achieved the best overall performance with an average accuracy of 96.89%, demonstrating the potential of data-driven deep architectures for reliable, single-trial C-VEP decoding in adaptive non-invasive BCI systems.},
}

@article {pmid41355286,
year = {2025},
author = {King, SE and Waddell, JT and Jan, I and McDonald, A and Raymond, C and Corbin, WR},
title = {Solitary drinking as a day-level risk factor for unique negative consequences among college students.},
journal = {Alcohol, clinical & experimental research},
volume = {},
number = {},
pages = {},
doi = {10.1111/acer.70216},
pmid = {41355286},
issn = {2993-7175},
support = {T32-DA039772/DA/NIDA NIH HHS/United States ; },
abstract = {BACKGROUND: Solitary drinking represents a high-risk pattern of drinking across individuals but when examined within individuals, solitary moments are associated with less risk. One possibility is that solitary drinking confers risk for specific negative consequences at the day level, but aggregate measures of negative consequences mask such relations. Thus, this study examined the extent to which solitary drinking increased the likelihood of reporting specific negative consequences, controlling for drinking quantity.

METHOD: College students (N = 1043; 51.8% female) completed a 30-day Timeline Followback Interview in which they reported day-level drinking context, drinking quantity, and negative consequences. A total of 7340 drinking days were reported. Two-level multilevel probit regressions with Bayesian estimation tested whether drinking context (i.e., solitary vs. social) was associated with an increased likelihood of experiencing each of eight unique negative consequences (i.e., social/interpersonal, risky behavior, blackouts, occupational, impaired control, physical dependence, self-care, and self-perception), controlling for drinking quantity.

RESULTS: When controlling for drinking quantity, solitary (vs. social) drinking days were associated with a higher likelihood of occupational consequences [β = 0.05, 95% BCI = (0.01, 0.08)] and diminished self-perception [β = 0.06, 95% BCI = (0.03, 0.10)]. Solitary drinking days were also associated with a lower likelihood of interpersonal consequences (β = -0.06, 95% BCI = [-0.11, -0.03]) and blackout drinking (β = -0.06, 95% BCI = [-0.09, -0.03]). Person-level results suggest that those who more often drink alone experience greater blackout drinking, impaired control, dependence, occupational consequences, and diminished self-perception (all p's < 0.001). When consequences were summed, solitary drinking days (vs. social) were associated with fewer negative consequences (β = -0.023, 95% BCI = [-0.049, -0.005]), whereas at the person level, those who more frequently drink alone experienced more negative consequences (β = 0.10, 95% BCI = [0.04, 0.17]).

CONCLUSIONS: Results suggest that social and solitary drinking contexts confer risk for specific consequences and that risk for consequences differs if consequences are aggregated. Findings may inform future interventions by emphasizing certain protective behavioral strategies in specific drinking contexts to reduce the likelihood of negative outcomes.},
}

@article {pmid41353180,
year = {2025},
author = {Soriano-Segura, P and Ortiz, M and Polo-Hortigüela, C and Iáñez, E and Azorín, JM},
title = {Characterization of error-related potentials during the command of a lower-limb exoskeleton based on deep learning.},
journal = {Journal of neuroengineering and rehabilitation},
volume = {},
number = {},
pages = {},
doi = {10.1186/s12984-025-01833-3},
pmid = {41353180},
issn = {1743-0003},
support = {PID2021-124111OB-C31//MICIU/AEI/10.13039/501100011033 and by ERDF, EU/ ; PRE2022-103336//MICIU/AEI/10.13039/501100011033 and by ERDF, EU/ ; },
}

@article {pmid41352637,
year = {2025},
author = {Roc, A and Kolodzienski, L and Dreyer, P and Appriou, A and Monseigne, T and Lotte, F},
title = {Evolution of users' subjective experience over three training sessions with an EEG Motor-Imagery Brain-Computer Interface (MI-BCI).},
journal = {Brain research},
volume = {},
number = {},
pages = {150085},
doi = {10.1016/j.brainres.2025.150085},
pmid = {41352637},
issn = {1872-6240},
abstract = {Motor Imagery-based Brain-Computer Interfaces (MI-BCIs) have been shown to be promising for numerous applications, including sport training and entertainment for healthy users, but also for improving or restoring functions in neurological and neuropsychiatric disorders, e.g., for motor rehabilitation post-stroke or for attention training in attention deficits. Reliable interactions with such MI-BCIs require a heavy training process for both the machine and the user. Yet, how User eXperience (UX) evolves during standard training is still largely unclear, both within and between sessions/days. Through an exploratory study, we investigated the variations of users' answers to a UX questionnaire when training with a standard left vs. right-hand MI-BCI. 24 healthy novice users engaged in 3 training sessions (with 12 runs each) on different days. Each short run was followed by six questions on screen measuring UX factors on scales from 1 to 10: mental demand, performance, mental effort, frustration, mental fatigue and anxiety. Interestingly, BCI performances did not correlate with any subjective UX measure in this study. However, a time effect was observed. Within session, the results suggested that mental demand, effort, and fatigue significantly augmented during BCI operation, and that frustration significantly fluctuated but did not differ pre- vs. post-session. Between sessions, the first session was rated significantly more challenging than the other two regarding frustration, anxiety, mental demand, mental effort and mental fatigue. This highlights the importance of conducting studies across sessions and of considering the users' mental states during BCI use, for improving UX and thus possibly BCI treatment outcome.},
}

@article {pmid41351188,
year = {2025},
author = {Wang, N and Chai, X and Song, J and He, Y and He, Q and Zhang, T and Liu, D and Li, J and Cao, T and Zhu, S and Jia, Y and Si, J and Ma, W and Yang, Y and Zhao, J},
title = {Motor Intention Quantization for Patients With Disorders of Consciousness by Multimodal BCI Combining Electroencephalography and Functional Near-Infrared Spectroscopy.},
journal = {CNS neuroscience & therapeutics},
volume = {31},
number = {12},
pages = {e70679},
doi = {10.1002/cns.70679},
pmid = {41351188},
issn = {1755-5949},
support = {7232049//Natural Science Foundation of Beijing Municipality/ ; 7252004//Natural Science Foundation of Beijing Municipality/ ; 2019-I2M-5-021//CAMS Innovation Fund for Medical Sciences（CIFMS）/ ; Z221100002722014//International (Hong Kong, Macao, and Taiwan) Science and Technology Cooperation Project/ ; 2022GKZS0003//2022 Open Project of Key Laboratory and Engineering Technology Research Center in the Rehabilitation Field of the Ministry of Civil Affairs/ ; 2022ZD0205300//Science and Technology Innovation 2030/ ; 82371197//National Natural Science Foundation of China/ ; 82501457//National Natural Science Foundation of China/ ; 2025-PUMCH-D-004//National High Level Hospital Clinical Research Funding/ ; },
mesh = {Humans ; Spectroscopy, Near-Infrared/methods ; *Electroencephalography/methods ; Female ; Male ; *Brain-Computer Interfaces ; Middle Aged ; *Consciousness Disorders/physiopathology/diagnosis ; Adult ; *Intention ; Aged ; Young Adult ; },
abstract = {OBJECTIVE: The current application of single-modality electroencephalography (EEG) or functional near-infrared spectroscopy (fNIRS) to assess consciousness levels in patients with disorders of consciousness (DoC) has garnered significant attention. However, the diagnostic accuracy of unimodal approaches remains suboptimal. Therefore, this study aims to apply the multimodal fusion technology of EEG and fNIRS to the clinical diagnosis of DoC patients.

METHODS: Eleven patients with DoC (six with a minimally conscious state [MCS] and five with a vegetative state [VS]) were enrolled. The motor intention-based brain-computer interface (MI-BCI) paradigm was adopted. EEG and fNIRS were recorded simultaneously. The synchronous states of EEG and fNIRS were analyzed, including time-frequency analysis, event-related desynchronization (ERD), and changes in oxy-hemoglobin (HbO)/de-oxygenated (HbR)/total hemoglobin (HbT) content. A multimodal method combining EEG and fNIRS was used to classify DoC patients.

RESULTS: The machine-learning results of the MI-BCI model showed that the EEG-fNIRS multimodal approach was superior to single-modality techniques in the diagnosis of healthy controls (HC), MCS, and VS. The multimodal model achieved a mean AUC of 0.69 ± 0.10, significantly outperforming both unimodal EEG (0.43 ± 0.19; p < 0.01) and standalone fNIRS (0.63 ± 0.10; p < 0.05). The EEG_ERD index of left-handed MI-BCI significantly differentiated the MCS and VS groups. Meanwhile, for the classification tasks of HC, MCS, and VS, the importance ranking of the indicators was as follows: fNIRS_ACC, EEG_ACC, fNIRS_slope, fNIRS_centroid, EEG_ERD, fNIRS_integral, and fNIRS_mean.

CONCLUSION: The integration of multimodal MI-BCI paradigms demonstrates clinical potential in evaluating consciousness levels, while the synergistic combination of neurophysiological and hemodynamic biomarkers provides a robust framework for enhancing the precision of bedside diagnostic protocols.

TRIAL REGISTRATION: Clinical Trial Registry: ChiCTR2400085830.},
}

Humans
Spectroscopy, Near-Infrared/methods
*Electroencephalography/methods
Female
Male
*Brain-Computer Interfaces
Middle Aged
*Consciousness Disorders/physiopathology/diagnosis
Adult
*Intention
Aged
Young Adult

@article {pmid41351118,
year = {2025},
author = {Liu, Q and Zhang, X and Zhang, H and Chen, K and He, Y and Niu, J and Li, W and Chen, H and Zhang, D and Li, J and Liao, W},
title = {Same movies, different stories: aberrant brain state dynamics during naturalistic emotional stimuli in depression.},
journal = {Journal of translational medicine},
volume = {},
number = {},
pages = {},
doi = {10.1186/s12967-025-07512-0},
pmid = {41351118},
issn = {1479-5876},
}

@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