@article {pmid39808931,
year = {2025},
author = {Russo, JS and Shiels, TA and Lin, CS and John, SE and Grayden, DB},
title = {Decoding imagined movement in people with multiple sclerosis for brain-computer interface translation.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/adaa1d},
pmid = {39808931},
issn = {1741-2552},
abstract = {Multiple Sclerosis (MS) is a heterogeneous autoimmune-mediated disorder affecting the central nervous system, commonly manifesting as fatigue and progressive limb impairment. This can significantly impact quality of life due to weakness or paralysis in the upper and lower limbs. A Brain-Computer Interface (BCI) aims to restore quality of life through control of an external device, such as a wheelchair. However, the limited BCI research in people with MS has been confined to exploring the P300 response and brain signals associated with attempted movement. The current study aims to expand the MS-BCI literature by highlighting the feasibility of decoding MS imagined movement. Approach. We collected electroencephalography (EEG) data from eight participants with various symptoms of MS and ten neurotypical control participants. Participants made imagined movements of the hands and feet as directed by a go no-go protocol. Binary regularised linear discriminant analysis was used to classify imagined movement vs. rest and vs. movement at individual time-frequency points. The frequency bands which provided the maximal accuracy, and the associated latency, were compared. Main Results. In all MS participants, the classification algorithm achieved above 70% accuracy in at least one imagined movement vs. rest classification and most movement vs. movement classifications. There was no significant difference between classification of limbs with weakness or paralysis to neurotypical controls. Both the MS and control groups possessed decodable information within the alpha (7-13 Hz) and beta (16-30 Hz) bands at similar latency. Significance. This study is the first to demonstrate the feasibility of decoding imagined movements in people with MS. As an alternative to the P300 response, motor imagery-based control of a BCI may also be combined with existing motor imagery therapy to supplement MS rehabilitation. These promising results merit further long term BCI studies to investigate the effect of MS progression on classification performance. .},
}

@article {pmid39808922,
year = {2025},
author = {Dekleva, BM and Collinger, J},
title = {Using transient, effector-specific neural responses to gate decoding for brain-computer interfaces.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/adaa1f},
pmid = {39808922},
issn = {1741-2552},
abstract = {Real-world implementation of brain-computer interfaces (BCI) for continuous control of devices should ideally rely on fully asynchronous decoding approaches. That is, the decoding algorithm should continuously update its output by estimating the user's intended actions from real-time neural activity, without the need for any temporal alignment to an external cue. This kind of open-ended temporal flexibility is necessary to achieve naturalistic and intuitive control, but presents a challenge: how do we know when it is appropriate to decode anything at all? Activity in motor cortex is dynamic and modulates with many different types of actions (proximal arm control, hand control, speech, etc.), which can interfere with each other. Additionally, the "decodability" of any given action type (amount of relevant information present in the neural activity) fluctuates over time based on motor intent as well as intrinsic network dynamics. Here we present a method for simplifying the problem of continual decoding that uses transient, end effector-specific neural responses to identify periods of effector engagement. For example, we have observed unique neural signatures at the onset and offset of hand-related actions. Only after detecting the period of engagement do we then decode specific action features (e.g. digit movement or force). By using this gated approach, decoding models can be simpler (owing to local linearities) and are less sensitive to interference from cross-effector interference such as combined reaching and grasping actions. Clinical Trial ID: NCT01894802.},
}

@article {pmid39809040,
year = {2025},
author = {Xu, R and Allison, BZ and Zhao, X and Liang, W and Wang, X and Cichocki, A and Jin, J},
title = {Multi-Scale Pyramid Squeeze Attention Similarity Optimization Classification Neural Network for ERP Detection.},
journal = {Neural networks : the official journal of the International Neural Network Society},
volume = {184},
number = {},
pages = {107124},
doi = {10.1016/j.neunet.2025.107124},
pmid = {39809040},
issn = {1879-2782},
abstract = {Event-related potentials (ERPs) can reveal brain activity elicited by external stimuli. Innovative methods to decode ERPs could enhance the accuracy of brain-computer interface (BCI) technology and promote the understanding of cognitive processes. This paper proposes a novel Multi-Scale Pyramid Squeeze Attention Similarity Optimization Classification Neural Network (MS-PSA-SOC) for ERP Detection. The model integrates a multi-scale architecture, self-attention mechanism, and deep metric learning to achieve a more comprehensive, refined, and discriminative feature representation. The MS module aggregates fine-grained local features and global features with a larger receptive field within a multi-scale architecture, effectively capturing the dynamic characteristics of complex oscillatory activities in the brain at different levels of abstraction. This preserves complementary spatiotemporal representation information. The PSA module continues the multi-scale contextual modeling from the previous module and achieves adaptive recalibration of multi-scale features. By employing effective aggregation and selection mechanisms, it highlights key features while suppressing redundant information. The SOC module jointly optimizes similarity metric loss and classification loss, maintaining the feature space distribution while focusing on sample class labels. This optimization of similarity relationships between samples improves the model's generalization ability and robustness. Results from public and self-collected datasets demonstrate that the command recognition accuracy of the MS-PSA-SOC model is at least 3.1% and 2.8% higher than other advanced algorithms, achieving superior performance. Additionally, the method demonstrates a lower standard deviation across both datasets. This study also validated the network parameters based on Shannon's sampling theorem and EEG "microstates" through relevant experiments.},
}

@article {pmid39808940,
year = {2025},
author = {Gu, M and Pei, W and Gao, X and Wang, Y},
title = {Optimizing the proportion of stimulation area in a grid stimulus for user-friendly SSVEP-based BCIs.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/adaa1e},
pmid = {39808940},
issn = {1741-2552},
abstract = {OBJECTIVE: Steady-state visual evoked potentials (SSVEPs) rely on the photic driving response to encode electroencephalogram (EEG) signals stably and efficiently. However, the user experience of the traditional stimulation with high-contrast flickers urgently needs to be improved. In this study, we introduce a novel paradigm of grid stimulation with weak flickering perception, distinguished by a markedly lower proportion of stimulation area in the overall pattern.

APPROACH: In an offline single-target experiment, we investigated the unique characteristics of SSVEPs evoked by varying proportions in grid stimuli within low and medium frequency bands. Based on the analysis of simulation performance across a four-class brain-computer interface (BCI) task and the evaluation of user experience questionnaires, a subset of paradigms that balance performance and comfort were selected for implementation in four-target online BCI systems.

MAIN RESULTS: Our results demonstrate that even ultra-low stimulation proportion paradigms can still evoke strong responses within specific frequency bands, effectively enhancing user experience with low and middle frequency stimuli. Notably, proportions of 0.94% and 2.10% within the 3-5 Hz range provide an optimal balance between performance and user experience. For frequencies extending up to 15 Hz, a 2.10% proportion remains ideal. At 20 Hz, slightly higher proportions of 3.75% and 8.43% maintain these benefits.

SIGNIFICANCE: These findings are crucial for advancing the development of effective and user-friendly SSVEP-based BCI systems.},
}

@article {pmid39808939,
year = {2025},
author = {Prakash, P and Lei, T and Flint, RD and Hsieh, JK and Fitzgerald, Z and Mugler, EM and Templer, J and Goldrick, MA and Tate, MC and Rosenow, JM and Glaser, JI and Slutzky, MW},
title = {Decoding speech intent from non-frontal cortical areas.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/adaa20},
pmid = {39808939},
issn = {1741-2552},
abstract = {Brain-machine interfaces (BMIs) have advanced greatly in decoding speech signals originating from the speech motor cortices. Primarily, these BMIs target individuals with intact speech motor cortices but who are paralyzed by disrupted connections between frontal cortices and their articulators due to brainstem stroke or motor neuron diseases such as amyotrophic lateral sclerosis. A few studies have shown some information outside the speech motor cortices, such as in parietal and temporal lobes, that also may be useful for BMIs. The ability to use information from outside the frontal lobes could be useful not only for people with locked-in syndrome but also to people with frontal lobe damage, which can cause nonfluent aphasia or apraxia of speech. However, temporal and parietal lobes are predominantly involved in perceptive speech processing and comprehension. Therefore, to be able to use signals from these areas in a speech BMI, it is important to ascertain that they are related to production. Here, using intracranial recordings, we sought evidence for whether, when and where neural information related to speech intention could be found in the temporal and parietal cortices. Causal information enabled us to distinguish speech intent from resting state and other processes involved in language processing or working memory. Information related to speech intent was distributed widely across the temporal and parietal lobes, including superior temporal, medial temporal, angular, and supramarginal gyri. This provides evidence of a decodable production-related signal in these areas. This insight may help in designing speech brain-machine interfaces that could benefit people with locked-in syndrome, aphasia or apraxia of speech.},
}

@article {pmid39805390,
year = {2025},
author = {Xu, A and Huang, Y and Wu, B and Zhang, J and Deng, B and Cai, M and Cao, J and Wang, J and Yang, B and Shao, X and He, Q and Ying, M},
title = {Phase separation-based screening identifies arsenic trioxide as the N-Myc-DNA interaction inhibitor for neuroblastoma therapy.},
journal = {Cancer letters},
volume = {},
number = {},
pages = {217449},
doi = {10.1016/j.canlet.2025.217449},
pmid = {39805390},
issn = {1872-7980},
}

@article {pmid39805258,
year = {2025},
author = {Liu, DH and Kumar, S and Alawieh, H and Racz, FS and Millan, JDR},
title = {Personalized μ-transcranial alternating current stimulation improves online brain-computer interface control.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/ada9c0},
pmid = {39805258},
issn = {1741-2552},
abstract = {OBJECTIVE: A motor imagery (MI)-based brain-computer interface (BCI) enables users to engage with external environments by capturing and decoding electroencephalography (EEG) signals associated with the imagined movement of specific limbs. Despite significant advancements in BCI technologies over the past 40 years, a notable challenge remains: many users lack BCI proficiency, unable to produce sufficiently distinct and reliable MI brain patterns, hence leading to low classification rates in their BCIs. The objective of this study is to enhance the online performance of MI-BCIs in a personalized, biomarker-driven approach using transcranial alternating current stimulation (tACS).

APPROACH: Previous studies have identified that the peak power spectral density (PSD) value in sensorimotor idling rhythms is a neural correlate of participants' upper limb MI-BCI performances. In this active-controlled, single-blind study, we applied 20 minutes of tACS at the participant-specific, peak µ frequency in resting-state sensorimotor rhythms (SMRs), with the goal of enhancing resting-state µ SMRs.

MAIN RESULTS: After tACS, we observed significant improvements in event-related desynchronizations (ERDs) of µ sensorimotor rhythms (SMRs), and in the performance of an online MI-BCI that decodes left versus right hand commands in healthy participants (N=10) -but not in an active control-stimulation control group (N=10). Lastly, we showed a significant correlation between the resting-state µ SMRs and µ ERD, offering a mechanistic interpretation behind the observed changes in online BCI performances.

SIGNIFICANCE: Our research lays the groundwork for future non-invasive interventions designed to enhance BCI performances, thereby improving the independence and interactions of individuals who rely on these systems.},
}

@article {pmid39803338,
year = {2024},
author = {Kim, KH and Jeong, JH and Ko, MJ and Lee, S and Kwon, WK and Lee, BJ},
title = {Using Artificial Intelligence in the Comprehensive Management of Spinal Cord Injury.},
journal = {Korean journal of neurotrauma},
volume = {20},
number = {4},
pages = {215-224},
pmid = {39803338},
issn = {2234-8999},
abstract = {Spinal cord injury (SCI) frequently results in persistent motor, sensory, or autonomic dysfunction, and the outcomes are largely determined by the location and severity of the injury. Despite significant technological progress, the intricate nature of the spinal cord anatomy and the difficulties associated with neuroregeneration make full recovery from SCI uncommon. This review explores the potential of artificial intelligence (AI), with a particular focus on machine learning, to enhance patient outcomes in SCI management. The application of AI, specifically machine learning, has revolutionized the diagnosis, treatment, prognosis, and rehabilitation of patients with SCI. By leveraging large datasets and identifying complex patterns, AI contributes to improved diagnostic accuracy, optimizes surgical procedures, and enables the personalization of therapeutic interventions. AI-driven prognostic models provide accurate predictions of recovery, facilitating improved planning and resource allocation. Additionally, AI-powered rehabilitation systems, including robotic devices and brain-computer interfaces, increase the effectiveness and accessibility of therapy. However, realizing the full potential of AI in SCI care requires ongoing research, interdisciplinary collaboration, and the development of comprehensive datasets. As AI continues to evolve, it is expected to play an increasingly vital role in enhancing the outcomes of patients with SCI.},
}

@article {pmid39801915,
year = {2025},
author = {Gu, C and Jin, X and Zhu, L and Yi, H and Liu, H and Yang, X and Babiloni, F and Kong, W},
title = {Cross-session SSVEP brainprint recognition using attentive multi-sub-band depth identity embedding learning network.},
journal = {Cognitive neurodynamics},
volume = {19},
number = {1},
pages = {15},
pmid = {39801915},
issn = {1871-4080},
abstract = {Brainprint recognition technology, regarded as a promising biometric technology, encounters challenges stemming from the time-varied, low signal-to-noise ratio of brain signals, such as electroencephalogram (EEG). Steady-state visual evoked potentials (SSVEP) exhibit high signal-to-noise ratio and frequency locking, making them a promising paradigm for brainprint recognition. Consequently, the extraction of time-invariant identity information from SSVEP EEG signals is essential. In this paper, we propose an Attentive Multi-sub-band Depth Identity Embedding Learning Network for stable cross-session SSVEP brainprint recognition. To address the issue of low recognition accuracy across sessions, we introduce the Sub-band Attentive Frequency mechanism, which integrates the frequency-domain relevant characteristics of the SSVEP paradigm and focuses on exploring depth-frequency identity embedding information. Also, we employ Attentive Statistic Pooling to enhance the stability of frequency domain feature distributions across sessions. Extensive experimentation and validation were conducted on two multi-session SSVEP benchmark datasets. The experimental results show that our approach outperforms other state-of-art models on 2-second samples across sessions and has the potential to serve as a benchmark in multi-subject biometric recognition systems.},
}

@article {pmid39801913,
year = {2025},
author = {Zhou, Y and Wang, P and Gong, P and Wan, P and Wen, X and Zhang, D},
title = {Cross-subject mental workload recognition using bi-classifier domain adversarial learning.},
journal = {Cognitive neurodynamics},
volume = {19},
number = {1},
pages = {16},
pmid = {39801913},
issn = {1871-4080},
abstract = {To deploy Electroencephalogram (EEG) based Mental Workload Recognition (MWR) systems in the real world, it is crucial to develop general models that can be applied across subjects. Previous studies have utilized domain adaptation to mitigate inter-subject discrepancies in EEG data distributions. However, they have focused on reducing global domain discrepancy, while neglecting local workload-categorical domain divergence. This degrades the workload-discriminating ability of subject-invariant features. To deal with this problem, we propose a novel joint category-wise and domain-wise alignment Domain Adaptation (cdaDA) algorithm, using bi-classifier learning and domain discriminative adversarial learning. The bi-classifier learning approach is adopted to address the similarities and differences between categories, helping to align EEG data within the same mental workload categories. Additionally, the domain discriminative adversarial learning technique is adopted to consider global domain information by minimizing global domain discrepancy. By integrating both local category information and global domain information, the cdaDA model performs a coarse-to-fine alignment and achieves promising cross-subject MWR results.},
}

@article {pmid39801910,
year = {2025},
author = {Wang, J and Zhang, L and Chen, S and Xue, H and Du, M and Xu, Y and Liu, S and Ming, D},
title = {Individuals with high autistic traits exhibit altered interhemispheric brain functional connectivity patterns.},
journal = {Cognitive neurodynamics},
volume = {19},
number = {1},
pages = {9},
pmid = {39801910},
issn = {1871-4080},
abstract = {Individuals with high autistic traits (AT) encounter challenges in social interaction, similar to autistic persons. Precise screening and focused interventions positively contribute to improving this situation. Functional connectivity analyses can measure information transmission and integration between brain regions, providing neurophysiological insights into these challenges. This study aimed to investigate the patterns of brain networks in high AT individuals to offer theoretical support for screening and intervention decisions. EEG data were collected during a 4-min resting state session with eyes open and closed from 48 participants. Using the Autism Spectrum Quotient (AQ) scale, participants were categorized into the high AT group (HAT, n = 15) and low AT groups (LAT, n = 15). We computed the interhemispheric and intrahemispheric alpha coherence in two groups. The correlation between physiological indices and AQ scores was also examined. Results revealed that HAT exhibited significantly lower alpha coherence in the homologous hemispheres of the occipital cortex compared to LAT during the eyes-closed resting state. Additionally, significant negative correlations were observed between the degree of AT (AQ scores) and the alpha coherence in the occipital cortex, as well as in the right frontal and left occipital regions. The findings indicated that high AT individuals exhibit decreased connectivity in the occipital region, potentially resulting in diminished ability to process social information from visual inputs. Our discovery contributes to a deeper comprehension of the neural underpinnings of social challenges in high AT individuals, providing neurophysiological signatures for screening and intervention strategies for this population.},
}

@article {pmid39798828,
year = {2025},
author = {Wu, L and Jiang, M and Zhao, M and Hu, X and Wang, J and Zhang, K and Jia, K and Ren, F and Gao, F},
title = {Right Inferior Frontal Cortex and preSMA in Response Inhibition: An Investigation Based on PTC Model.},
journal = {NeuroImage},
volume = {},
number = {},
pages = {121004},
doi = {10.1016/j.neuroimage.2025.121004},
pmid = {39798828},
issn = {1095-9572},
abstract = {Response inhibition is an essential component of cognitive function. A large body of literature has used neuroimaging data to uncover the neural architecture that regulates inhibitory control in general and movement cancelation. The presupplementary motor area (preSMA) and the right inferior frontal cortex (rIFC) are the key nodes in the inhibitory control network. However, how these two regions contribute to response inhibition remains controversial. Based on the Pause-then-Cancel Model (PTC), this study employed functional magnetic resonance imaging (fMRI) to investigate the functional specificity of two regions in the stopping process. The Go/No-Go task (GNGT) and the Stop Signal Task (SST) were administered to the same group of participants. We used the GNGT to dissociate the pause process and both the GNGT and the SST to investigate the inhibition mechanism. Imaging data revealed that response inhibition produced by both tasks activated the preSMA and rIFC. Furthermore, an across-participants analysis showed that increased activation in the rIFC was associated with a delay in the go response in the GNGT. In contrast, increased activation in the preSMA was associated with good inhibition efficiency via the striatum in both GNGT and SST. These behavioral and imaging findings support the PTC model of the role of rIFC and preSMA, that the former is involved in a pause process to delay motor responses, whereas the preSMA is involved in the stopping of motor responses.},
}

@article {pmid39796947,
year = {2024},
author = {Chio, N and Quiles-Cucarella, E},
title = {A Bibliometric Review of Brain-Computer Interfaces in Motor Imagery and Steady-State Visually Evoked Potentials for Applications in Rehabilitation and Robotics.},
journal = {Sensors (Basel, Switzerland)},
volume = {25},
number = {1},
pages = {},
doi = {10.3390/s25010154},
pmid = {39796947},
issn = {1424-8220},
mesh = {*Brain-Computer Interfaces ; Humans ; *Robotics/methods ; *Evoked Potentials, Visual/physiology ; *Bibliometrics ; Imagination/physiology ; Rehabilitation/methods ; Electroencephalography/methods ; },
abstract = {In this paper, a bibliometric review is conducted on brain-computer interfaces (BCI) in non-invasive paradigms like motor imagery (MI) and steady-state visually evoked potentials (SSVEP) for applications in rehabilitation and robotics. An exploratory and descriptive approach is used in the analysis. Computational tools such as the biblioshiny application for R-Bibliometrix and VOSViewer are employed to generate data on years, sources, authors, affiliation, country, documents, co-author, co-citation, and co-occurrence. This article allows for the identification of different bibliometric indicators such as the research process, evolution, visibility, volume, influence, impact, and production in the field of brain-computer interfaces for MI and SSVEP paradigms in rehabilitation and robotics applications from 2000 to August 2024.},
}

*Brain-Computer Interfaces
Humans
*Robotics/methods
*Evoked Potentials, Visual/physiology
*Bibliometrics
Imagination/physiology
Rehabilitation/methods
Electroencephalography/methods

@article {pmid39796911,
year = {2024},
author = {Özkahraman, A and Ölmez, T and Dokur, Z},
title = {Performance Improvement with Reduced Number of Channels in Motor Imagery BCI System.},
journal = {Sensors (Basel, Switzerland)},
volume = {25},
number = {1},
pages = {},
doi = {10.3390/s25010120},
pmid = {39796911},
issn = {1424-8220},
mesh = {*Brain-Computer Interfaces ; Humans ; *Electroencephalography/methods ; Electrooculography/methods ; Deep Learning ; Signal Processing, Computer-Assisted ; Algorithms ; },
abstract = {Classifying Motor Imaging (MI) Electroencephalogram (EEG) signals is of vital importance for Brain-Computer Interface (BCI) systems, but challenges remain. A key challenge is to reduce the number of channels to improve flexibility, portability, and computational efficiency, especially in multi-class scenarios where more channels are needed for accurate classification. This study demonstrates that combining Electrooculogram (EOG) channels with a reduced set of EEG channels is more effective than relying on a large number of EEG channels alone. EOG channels provide useful information for MI signal classification, countering the notion that they only introduce eye-related noise. The study uses advanced deep learning techniques, including multiple 1D convolution blocks and depthwise-separable convolutions, to optimize classification accuracy. The findings in this study are tested on two datasets: dataset 1, the BCI Competition IV Dataset IIa (4-class MI), and dataset 2, the Weibo dataset (7-class MI). The performance for dataset 1, utilizing 3 EEG and 3 EOG channels (6 channels total), is of 83% accuracy, while dataset 2, with 3 EEG and 2 EOG channels (5 channels total), achieves an accuracy of 61%, demonstrating the effectiveness of the proposed channel reduction method and deep learning model.},
}

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

@article {pmid39793200,
year = {2025},
author = {Peters, B and Celik, B and Gaines, D and Galvin-McLaughlin, D and Imbiriba, T and Kinsella, M and Klee, D and Lawhead, M and Memmott, T and Smedemark-Margulies, N and Wiedrick, J and Erdogmus, D and Oken, B and Vertanen, K and Fried-Oken, M},
title = {RSVP Keyboard with Inquiry Preview: mixed performance and user experience with an adaptive, multimodal typing interface combining EEG and switch input.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/ada8e0},
pmid = {39793200},
issn = {1741-2552},
abstract = {OBJECTIVE: The RSVP Keyboard is a non-implantable, event-related potential-based brain-computer interface (BCI) system designed to support communication access for people with severe speech and physical impairments. Here we introduce Inquiry Preview, a new RSVP Keyboard interface incorporating switch input for users with some voluntary motor function, and describe its effects on typing performance and other outcomes.

APPROACH: Four individuals with disabilities participated in the collaborative design of possible switch input applications for the RSVP Keyboard, leading to the development of Inquiry Preview and a method of fusing switch input with language model and electroencephalography (EEG) evidence for typing. Twenty-four participants without disabilities and one potential end user with incomplete locked-in syndrome took part in two experiments investigating the effects of Inquiry Preview and two modes of switch input on typing accuracy and speed during a copy-spelling task.

MAIN RESULTS: For participants without disabilities, Inquiry Preview and switch input tended to worsen typing performance compared to the standard RSVP Keyboard condition, with more consistent effects across participants for speed than for accuracy. However, there was considerable variability, with some participants demonstrating improved typing performance and better user experience with Inquiry Preview and switch input. Typing performance for the potential end user was comparable to that of participants without disabilities. He typed most quickly and accurately with Inquiry Preview and switch input and gave favorable user experience ratings to those conditions, but preferred standard RSVP Keyboard.

SIGNIFICANCE: Inquiry Preview is a novel multimodal interface for the RSVP Keyboard BCI, incorporating switch input as an additional control signal. Typing performance and user experience and preference varied widely across participants, reinforcing the need for flexible, customizable BCI systems that can adapt to individual users.

CLINICALTRIALS: gov Identifier: NCT04468919.},
}

@article {pmid39794127,
year = {2025},
author = {Chang, A and Poeppel, D and Teng, X},
title = {Temporally dissociable neural representations of pitch height and chroma.},
journal = {The Journal of neuroscience : the official journal of the Society for Neuroscience},
volume = {},
number = {},
pages = {},
doi = {10.1523/JNEUROSCI.1567-24.2024},
pmid = {39794127},
issn = {1529-2401},
abstract = {The extraction and analysis of pitch underpin speech and music recognition, sound segregation, and other auditory tasks. Perceptually, pitch can be represented as a helix composed of two factors: height monotonically aligns with frequency, while chroma cyclically repeats at doubled frequencies. Although the early perceptual and neurophysiological mechanisms for extracting pitch from acoustic signals have been extensively investigated, the equally essential subsequent stages that bridge to high-level auditory cognition remain less well understood. How does the brain represent perceptual attributes of pitch at higher-order processing stages and how are the neural representations formed over time? We used a machine learning approach to decode time-resolved neural responses of human listeners (10 females and 7 males) measured by magnetoencephalography across different pitches, hypothesizing that different pitches sharing similar neural representations would result in reduced decoding performance. We show that pitch can be decoded from lower-frequency neural responses within auditory-frontal cortical regions. Specifically, linear mixed-effect modeling reveals that height and chroma explain decoding performance of delta band (0.5-4 Hz) neural activity at distinct latencies: a long-lasting height effect precedes a transient chroma effect, followed by a recurrence of height after chroma, indicating sequential processing stages associated with unique perceptual and neural characteristics. Furthermore, the localization analyses of the decoder demonstrate that height and chroma are associated with overlapping cortical regions, with differences observed in the right orbital and polar frontal cortex. The data provide a perspective motivating new hypotheses on the mechanisms of pitch representation.Significance Statement Pitch is fundamental to various facets of human hearing, including music appreciation, speech comprehension, vocal learning, and sound source differentiation. How does the brain encode the perceptual features of pitch? By applying machine learning techniques to time-resolved neuroimaging data of individuals listening to different pitches, our findings reveal that pitch height and chroma-two distinct features of pitch-are associated with different neural dynamics within the auditory-frontal cortical network, with height playing a more prominent role. This offers a unified theoretical framework for understanding the perceptual and neural characteristics of pitch perception and opens new avenues for noninvasively decoding human auditory perception to develop brain-computer interfaces.},
}

@article {pmid39793091,
year = {2025},
author = {Zhang, F and Pu, Y and Kong, XZ},
title = {Parallel vector memories or single memory updating?.},
journal = {Proceedings of the National Academy of Sciences of the United States of America},
volume = {122},
number = {1},
pages = {e2422788121},
doi = {10.1073/pnas.2422788121},
pmid = {39793091},
issn = {1091-6490},
support = {32171031//MOST | National Natural Science Foundation of China (NSFC)/ ; 32400882//MOST | National Natural Science Foundation of China (NSFC)/ ; 2021ZD0200409//STI 2030-Major Projects/ ; },
}

@article {pmid39793054,
year = {2025},
author = {Chivukula, S and Aflalo, T and Zhang, C and Rosario, ER and Bari, A and Pouratian, N and Andersen, RA},
title = {Population encoding of observed and actual somatosensations in the human posterior parietal cortex.},
journal = {Proceedings of the National Academy of Sciences of the United States of America},
volume = {122},
number = {1},
pages = {e2316012121},
doi = {10.1073/pnas.2316012121},
pmid = {39793054},
issn = {1091-6490},
support = {R01EY015545//HHS | NIH | National Eye Institute (NEI)/ ; P50MH094258//CIT | Caltech Conte Center for Social Decision Making (Caltech Conte Center for Neuroscience)/ ; },
mesh = {Humans ; *Parietal Lobe/physiology ; Female ; Touch/physiology ; Cognition/physiology ; Adult ; Neurons/physiology ; Mirror Neurons/physiology ; Touch Perception/physiology ; Brain-Computer Interfaces ; },
abstract = {Cognition relies on transforming sensory inputs into a generalizable understanding of the world. Mirror neurons have been proposed to underlie this process, mapping visual representations of others' actions and sensations onto neurons that mediate our own, providing a conduit for understanding. However, this theory has limitations. Here, we hypothesize that mirror-like responses represent one facet of a broader framework in which our brains engage internal models for cognition. We recorded populations of single neurons in the human posterior parietal cortex (PPC) of a brain-machine interface clinical trial participant implanted with a microelectrode array while she either experienced actual touch, or observed diverse tactile stimuli applied to other individuals. Two body locations were tested, on each of the participant and other individuals. Some neurons exhibited mirror-like properties, consistent with earlier literature. However, they were fragile, breaking with increased task complexity. Population responses were better characterized by generalizable and compositional basic-level features encoded within neural subspaces. These features enable the population to respond to diverse actual and observed touch stimuli and are recruited similarly for similar forms of touch. Mirror-like neurons belong within these subspaces, contributing more globally to compositionality and generalizability. We speculate that at a population-level, human PPC manifests an internal model for touch, and that cognition unfolds in the high-level human cortex by versatility in its representational building blocks. In a broad sense, we speculate that the population features we demonstrate support a broad mechanism by which the high-level human cortex enables understanding.},
}

Humans
*Parietal Lobe/physiology
Female
Touch/physiology
Cognition/physiology
Adult
Neurons/physiology
Mirror Neurons/physiology
Touch Perception/physiology
Brain-Computer Interfaces

@article {pmid39789264,
year = {2025},
author = {Lin, N and Wang, S and Li, Y and Wang, B and Shi, S and He, Y and Zhang, W and Yu, Y and Zhang, Y and Zhang, X and Wong, K and Wang, S and Chen, X and Jiang, H and Zhang, X and Lin, P and Xu, X and Qi, X and Wang, Z and Shang, D and Liu, Q and Liu, M},
title = {Resistive memory-based zero-shot liquid state machine for multimodal event data learning.},
journal = {Nature computational science},
volume = {},
number = {},
pages = {},
pmid = {39789264},
issn = {2662-8457},
support = {62422004//National Natural Science Foundation of China (National Science Foundation of China)/ ; Z210006//Natural Science Foundation of Beijing Municipality (Beijing Natural Science Foundation)/ ; },
abstract = {The human brain is a complex spiking neural network (SNN) capable of learning multimodal signals in a zero-shot manner by generalizing existing knowledge. Remarkably, it maintains minimal power consumption through event-based signal propagation. However, replicating the human brain in neuromorphic hardware presents both hardware and software challenges. Hardware limitations, such as the slowdown of Moore's law and Von Neumann bottleneck, hinder the efficiency of digital computers. In addition, SNNs are characterized by their software training complexities. Here, to this end, we propose a hardware-software co-design on a 40 nm 256 kB in-memory computing macro that physically integrates a fixed and random liquid state machine SNN encoder with trainable artificial neural network projections. We showcase the zero-shot learning of multimodal events on the N-MNIST and N-TIDIGITS datasets, including visual and audio data association, as well as neural and visual data alignment for brain-machine interfaces. Our co-design achieves classification accuracy comparable to fully optimized software models, resulting in a 152.83- and 393.07-fold reduction in training costs compared with state-of-the-art spiking recurrent neural network-based contrastive learning and prototypical networks, and a 23.34- and 160-fold improvement in energy efficiency compared with cutting-edge digital hardware, respectively. These proof-of-principle prototypes demonstrate zero-shot multimodal events learning capability for emerging efficient and compact neuromorphic hardware.},
}

@article {pmid39788124,
year = {2025},
author = {An, D and You, Y and Ma, Q and Xu, Z and Liu, Z and Liao, R and Chen, H and Wang, Y and Wang, Y and Dai, H and Li, H and Jiang, L and Chen, Z and Hu, W},
title = {Deficiency of histamine H2 receptors in parvalbumin-positive neurons leads to hyperactivity, impulsivity, and impaired attention.},
journal = {Neuron},
volume = {},
number = {},
pages = {},
doi = {10.1016/j.neuron.2024.12.002},
pmid = {39788124},
issn = {1097-4199},
abstract = {Attention deficit hyperactivity disorder (ADHD), affecting 4% of the population, is characterized by inattention, hyperactivity, and impulsivity; however, its neurophysiological mechanisms remain unclear. Here, we discovered that deficiency of histamine H2 receptor (H2R) in parvalbumin-positive neurons in substantia nigra pars recticulata (PV[SNr]) attenuates PV[+] neuronal activity and induces hyperactivity, impulsivity, and inattention in mice. Moreover, decreased H2R expression was observed in PV[SNr] in patients with ADHD symptoms and dopamine-transporter-deficient mice, whose behavioral phenotypes were alleviated by H2R agonist treatment. Dysfunction of PV[SNr] efferents to the substantia nigra pars compacta dopaminergic neurons and superior colliculus differently contributes to H2R-deficiency-induced behavioral disorders. Collectively, our results demonstrate that H2R deficiency in PV[+] neurons contributes to hyperactivity, impulsivity, and inattention by dampening PV[SNr] activity and involving different efferents in mice. It may enhance understanding of the molecular and circuit-level basis of ADHD and afford new potential therapeutic targets for ADHD-like psychiatric diseases.},
}

@article {pmid39787896,
year = {2025},
author = {Li, L and Menendez-Lustri, DM and Hartzler, A and Pogharian, A and Zaorski, B and Chen, A and Palen, J and Traylor, B and Quill, E and Pawlowski, CL and Bruckman, MA and Gupta, AS and Capadona, JR and Shoffstall, AJ},
title = {Systemically administered platelet-inspired nanoparticles to reduce inflammation surrounding intracortical microelectrodes.},
journal = {Biomaterials},
volume = {317},
number = {},
pages = {123082},
doi = {10.1016/j.biomaterials.2025.123082},
pmid = {39787896},
issn = {1878-5905},
abstract = {Intracortical microelectrodes (IMEs) are essential for neural signal acquisition in neuroscience and brain-machine interface (BMI) systems, aiding patients with neurological disorders, paralysis, and amputations. However, IMEs often fail to maintain robust signal quality over time, partly due to neuroinflammation caused by vascular damage during insertion. Platelet-inspired nanoparticles (PIN), which possess injury-targeting functions, mimic the adhesion and aggregation of active platelets through conjugated collagen-binding peptides (CBP), von Willebrand Factor-binding peptides (VBP), and fibrinogen-mimetic peptides (FMP). Systemically administered PINs can potentially enhance hemostasis and promote the resealing of IME insertion-induced leaky blood-brain barrier (BBB), thereby attenuating the influx of blood-derived proteins into the brain parenchyma that trigger neuroinflammation. This study explores the potential of PINs to mitigate neuroinflammation at implant sites. Male Sprague Dawley rats underwent craniotomies and IME implantations, followed by a single dose of Cy5 labeled PINs (2 mg/kg). Rats were sacrificed at intervals from 0 to 4 days post-implantation (DPI) for biodistribution analysis using an in vivo live imaging system (IVIS) and immunohistochemistry (IHC) to assess neuroinflammation, BBB permeability, and active platelet distribution. Another cohort of rats received weekly PINs, trehalose buffer (TH, diluent control), or control nanoparticles (CP, PEG-coated liposomes) for 4 weeks, with similar endpoint analyses. Results indicated that PIN concentrations were significantly elevated near IME interfaces acutely (0-4 DPI) and after 4 weeks of repeated dosing. At 3 DPI, peak intensities of active platelets (CD62P), activated microglia/macrophages (CD68), and PINs were observed. Immunoglobulin G (IgG) was upregulated during the first 24 h near implant sites but declined thereafter. At 4 weeks, the PINs group exhibited higher intensities of active platelets and PINs, and reduced CD68 and IgG levels compared to controls. PINs effectively targeted the IME-tissue interface, alongside endogenous activated platelets, resulting in reduced neuroinflammatory and BBB-leakage markers compared to the diluent-only-infused control group. Repeated dosing of PINs presents a promising approach for enhancing the quality of neural recordings in future studies.},
}

@article {pmid39787745,
year = {2025},
author = {Elsohaby, I and Kostoulas, P and Fayez, M and Elmoslemany, A and Alkafafy, ME and Bahhary, AM and Alzahrani, R and Morsi, AEKM and Arango-Sabogal, JC},
title = {Bayesian estimation of diagnostic accuracy of fecal smears, fecal PCR and serum ELISA for detecting Mycobacterium avium subsp. paratuberculosis infections in four domestic ruminant species in Saudi Arabia.},
journal = {Veterinary microbiology},
volume = {301},
number = {},
pages = {110377},
doi = {10.1016/j.vetmic.2025.110377},
pmid = {39787745},
issn = {1873-2542},
abstract = {Paratuberculosis, a chronic wasting disease affecting domestic and wild ruminants worldwide, is caused by Mycobacterium avium subsp. paratuberculosis (MAP). Various diagnostic tests exist for detecting MAP infection; however, none of them possess perfect accuracy to be qualified as a reference standard test, particularly due to their notably low sensitivity. Therefore, we used Bayesian latent class models (BLCMs) to estimate diagnostic accuracy of fecal smears (FS), fecal PCR and serum ELISA for detecting MAP infections in sheep, goats, cattle, and camels older than 2 years in Saudi Arabia. Data from a cross-sectional study conducted in the Eastern Province of Saudi Arabia on 31 different farms with a history of MAP infection were analyzed. Fecal and blood samples from all animals older than 2 years in each farm were collected, resulting in a total of 220 sheep, 123 goats, 66 cattle, and 240 camels sampled. FS and IS900-PCR were performed on fecal samples to detect acid-fast bacilli and MAP DNA, respectively. The IDEXX ELISA kit was used to detect MAP antibodies in serum samples. For each ruminant species population, a BLCM was fitted to obtain posterior estimates [medians and 95 % Bayesian credible intervals (95 % BCI)] for sensitivity (Se) and specificity (Sp) of the three tests. We assumed FS and PCR to be conditionally dependent on the true animal MAP status. Prior distributions for test accuracy were used if available. FS had the highest Se among all tests and across all species with median values around 80 % in sheep, goats and camels, and near 50 % in cattle. Median Sp estimates of ELISA and PCR were higher than 90 % for all species. FS yielded the lowest Sp of the study when applied in camels, sheep, and goats. Using the prevalence observed in this study, median positive predictive value (PPV) was higher for PCR and ELISA than FS for camels, sheep, and goats. In cattle, PPV of all tests was similar with median estimates > 95 %. In camels, sheep, and goats, median negative predicative value (NPV) of all tests were > 60 %. The lowest median NPV for all tests were observed in cattle (< 30 %). Our results suggest that ELISA is a suitable option to identify MAP infected animals in farms with previous history of MAP in the Eastern region of Saudi Arabia.},
}

@article {pmid39787440,
year = {2024},
author = {Alkawadri, CI and Yan, Q and Kocuglu Kinal, AG and Spencer, DD and Alkawadri, R},
title = {Comparison of EEG Signal Characteristics of Subdural and Depth Electrodes.},
journal = {Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society},
volume = {},
number = {},
pages = {},
doi = {10.1097/WNP.0000000000001139},
pmid = {39787440},
issn = {1537-1603},
abstract = {OBJECTIVES: Our study aimed to compare signal characteristics of subdural electrodes (SDE) and depth stereo EEG placed within a 5-mm vicinity in patients with drug-resistant epilepsy. We report how electrode design and placement collectively affect signal content from a shared source between these electrode types.

METHODS: In subjects undergoing invasive intracranial EEG evaluation at a surgical epilepsy center from 2012 to 2018, stereo EEG and SDE electrode contacts placed within a 5-mm vicinity were identified. Of these, 24 contacts (12 pairs) met our criteria for signal-to-noise ratio and data availability for final analysis. We used Welch method to analyze the correlation of power spectral densities of EEG segments, root mean square of 1-second windows, and fast-Fourier transform to calculate coherence across conventional frequency bands.

RESULTS: We observed a median distance of 3.7 mm between the electrode contact pairs. Time-aware analysis highlighted the coherence's strength primarily in the high-gamma band, where the median (r) was 0.889. In addition, the median power ratios between the SDE and stereo EEG signal was 1.99. This ratio decreased from high-gamma to infra-low frequencies, with medians of 2.07 and 0.97, respectively. The power spectral densities for the stereo EEG and SDE electrodes demonstrated a strong correlation, with a median correlation coefficient (r) of 0.99 and an interquartile range from 0.915 to 0.996.

CONCLUSIONS: Signals captured by standard subdural and depth (intracranial EEG) electrodes within a 5-mm radius exhibit band-specific coherence and are not identical. The association was most pronounced in the high-gamma band, with coherence decreasing with lower frequencies. Our findings underscore the combined effects of electrode size, design, placement, preferred bandwidth, and the nature of the activity source on signal recording. Particularly, SDE employed herein may offer advantages for high-frequency signals, but the impact of electrode size on recordings necessitates careful consideration in context-specific situations.

SIGNIFICANCE: The findings relate to surgical epilepsy care and may inform the design of brain-computer interface.},
}

@article {pmid39781056,
year = {2025},
author = {Cai, Z and Li, P and Cheng, L and Yuan, D and Li, M and Li, H},
title = {A high performance heterogeneous hardware architecture for brain computer interface.},
journal = {Biomedical engineering letters},
volume = {15},
number = {1},
pages = {217-227},
pmid = {39781056},
issn = {2093-985X},
abstract = {Brain-computer interface (BCI) has been widely used in human-computer interaction. The introduction of artificial intelligence has further improved the performance of BCI system. In recent years, the development of BCI has gradually shifted from personal computers to embedded devices, which boasts lower power consumption and smaller size, but at the cost of limited device resources and computing speed, thus can hardly improve the support of complex algorithms. This paper proposes a heterogeneous BCI architecture based on ARM + FPGA, enabling real-time processing of electroencephalogram (EEG) signals. Adopting data quantization, layer fusion and data augmentation to optimize the compact neural network model EEGNet, and design dedicated hardware engines to accelerate the network. Experimental results show that the system achieves 93.3% classification accuracy for steady-state visual evoked potential signals, with a time delay of 0.2 ms per trail, and a power consumption of approximately (1.91 W). That is 31.5 times faster acceleration is realized at the cost of only 0.7% lower accuracy compared with the conventional processor. The results show that the BCI architecture proposed in this study has strong practicability and high research significance.},
}

@article {pmid39779796,
year = {2025},
author = {Shelishiyah, R and Thiyam, DB and Margaret, MJ and Banu, NMM},
title = {A hybrid CNN model for classification of motor tasks obtained from hybrid BCI system.},
journal = {Scientific reports},
volume = {15},
number = {1},
pages = {1360},
pmid = {39779796},
issn = {2045-2322},
mesh = {Humans ; *Brain-Computer Interfaces ; *Electroencephalography/methods ; *Neural Networks, Computer ; Deep Learning ; Adult ; Male ; Movement/physiology ; },
abstract = {The Hybrid-Brain Computer Interface (BCI) has shown improved performance, especially in classifying multi-class data. Two non-invasive BCI modules are combined to achieve an improved classification which are Electroencephalogram (EEG) and functional Near Infra-red Spectroscopy (fNIRS). Classifying contralateral and ipsilateral motor movements is found challenging among the other mental activity signals. The current work focuses on the performance of deep learning methods like - Convolutional Neural Networks (CNN) and Bidirectional Long-Short term memory (Bi-LSTM) in classifying a four-class motor execution of Right Hand, Left Hand, Right Arm and Left Arm taken from the CORE dataset. The model performance was evaluated using metrics such as Accuracy, F1 - score, Precision, Recall, AUC and ROC curve. The CNN and Hybrid CNN models have resulted in 98.3% and 99% accuracy respectively.},
}

Humans
*Brain-Computer Interfaces
*Electroencephalography/methods
*Neural Networks, Computer
Deep Learning
Adult
Male
Movement/physiology

@article {pmid39776784,
year = {2024},
author = {Ahmadi, S and Desain, P and Thielen, J},
title = {A Bayesian dynamic stopping method for evoked response brain-computer interfacing.},
journal = {Frontiers in human neuroscience},
volume = {18},
number = {},
pages = {1437965},
pmid = {39776784},
issn = {1662-5161},
abstract = {INTRODUCTION: As brain-computer interfacing (BCI) systems transition fromassistive technology to more diverse applications, their speed, reliability, and user experience become increasingly important. Dynamic stopping methods enhance BCI system speed by deciding at any moment whether to output a result or wait for more information. Such approach leverages trial variance, allowing good trials to be detected earlier, thereby speeding up the process without significantly compromising accuracy. Existing dynamic stopping algorithms typically optimize measures such as symbols per minute (SPM) and information transfer rate (ITR). However, these metrics may not accurately reflect system performance for specific applications or user types. Moreover, many methods depend on arbitrary thresholds or parameters that require extensive training data.

METHODS: We propose a model-based approach that takes advantage of the analytical knowledge that we have about the underlying classification model. By using a risk minimization approach, our model allows precise control over the types of errors and the balance between precision and speed. This adaptability makes it ideal for customizing BCI systems to meet the diverse needs of various applications.

RESULTS AND DISCUSSION: We validate our proposed method on a publicly available dataset, comparing it with established static and dynamic stopping methods. Our results demonstrate that our approach offers a broad range of accuracy-speed trade-offs and achieves higher precision than baseline stopping methods.},
}

@article {pmid39775950,
year = {2025},
author = {Zeng, X and Kang, T and Huang, W and Jin, T},
title = {How Should the BCI and BOOI Index be Correctly Applied in Patients With Low-Compliance Bladder?.},
journal = {Neurourology and urodynamics},
volume = {},
number = {},
pages = {},
doi = {10.1002/nau.25663},
pmid = {39775950},
issn = {1520-6777},
}

@article {pmid39771903,
year = {2024},
author = {Alzahrani, S and Banjar, H and Mirza, R},
title = {Systematic Review of EEG-Based Imagined Speech Classification Methods.},
journal = {Sensors (Basel, Switzerland)},
volume = {24},
number = {24},
pages = {},
pmid = {39771903},
issn = {1424-8220},
mesh = {Humans ; *Electroencephalography/methods ; *Brain-Computer Interfaces ; *Speech/physiology ; Machine Learning ; Signal Processing, Computer-Assisted ; Imagination/physiology ; Neural Networks, Computer ; Brain/physiology ; },
abstract = {This systematic review examines EEG-based imagined speech classification, emphasizing directional words essential for development in the brain-computer interface (BCI). This study employed a structured methodology to analyze approaches using public datasets, ensuring systematic evaluation and validation of results. This review highlights the feature extraction techniques that are pivotal to classification performance. These include deep learning, adaptive optimization, and frequency-specific decomposition, which enhance accuracy and robustness. Classification methods were explored by comparing traditional machine learning with deep learning and emphasizing the role of brain lateralization in imagined speech for effective recognition and classification. This study discusses the challenges of generalizability and scalability in imagined speech recognition, focusing on subject-independent approaches and multiclass scalability. Performance benchmarking across various datasets and methodologies revealed varied classification accuracies, reflecting the complexity and variability of EEG signals. This review concludes that challenges remain despite progress, particularly in classifying directional words. Future research directions include improved signal processing techniques, advanced neural network architectures, and more personalized, adaptive BCI systems. This review is critical for future efforts to develop practical communication tools for individuals with speech and motor impairments using EEG-based BCIs.},
}

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

@article {pmid39771862,
year = {2024},
author = {Khabti, J and AlAhmadi, S and Soudani, A},
title = {Enhancing Deep-Learning Classification for Remote Motor Imagery Rehabilitation Using Multi-Subject Transfer Learning in IoT Environment.},
journal = {Sensors (Basel, Switzerland)},
volume = {24},
number = {24},
pages = {},
pmid = {39771862},
issn = {1424-8220},
mesh = {Humans ; *Brain-Computer Interfaces ; *Deep Learning ; *Electroencephalography/methods ; Internet of Things ; Signal Processing, Computer-Assisted ; },
abstract = {One of the most promising applications for electroencephalogram (EEG)-based brain-computer interfaces (BCIs) is motor rehabilitation through motor imagery (MI) tasks. However, current MI training requires physical attendance, while remote MI training can be applied anywhere, facilitating flexible rehabilitation. Providing remote MI training raises challenges to ensuring an accurate recognition of MI tasks by healthcare providers, in addition to managing computation and communication costs. The MI tasks are recognized through EEG signal processing and classification, which can drain sensor energy due to the complexity of the data and the presence of redundant information, often influenced by subject-dependent factors. To address these challenges, we propose in this paper a multi-subject transfer-learning approach for an efficient MI training framework in remote rehabilitation within an IoT environment. For efficient implementation, we propose an IoT architecture that includes cloud/edge computing as a solution to enhance the system's efficiency and reduce the use of network resources. Furthermore, deep-learning classification with and without channel selection is applied in the cloud, while multi-subject transfer-learning classification is utilized at the edge node. Various transfer-learning strategies, including different epochs, freezing layers, and data divisions, were employed to improve accuracy and efficiency. To validate this framework, we used the BCI IV 2a dataset, focusing on subjects 7, 8, and 9 as targets. The results demonstrated that our approach significantly enhanced the average accuracy in both multi-subject and single-subject transfer-learning classification. In three-subject transfer-learning classification, the FCNNA model achieved up to 79.77% accuracy without channel selection and 76.90% with channel selection. For two-subject and single-subject transfer learning, the application of transfer learning improved the average accuracy by up to 6.55% and 12.19%, respectively, compared to classification without transfer learning. This framework offers a promising solution for remote MI rehabilitation, providing both accurate task recognition and efficient resource usage.},
}

Humans
*Brain-Computer Interfaces
*Deep Learning
*Electroencephalography/methods
Internet of Things
Signal Processing, Computer-Assisted

@article {pmid39771843,
year = {2024},
author = {Mikhaylov, D and Saeed, M and Husain Alhosani, M and F Al Wahedi, Y},
title = {Comparison of EEG Signal Spectral Characteristics Obtained with Consumer- and Research-Grade Devices.},
journal = {Sensors (Basel, Switzerland)},
volume = {24},
number = {24},
pages = {},
pmid = {39771843},
issn = {1424-8220},
mesh = {Humans ; *Electroencephalography/methods/instrumentation ; *Signal Processing, Computer-Assisted ; Adult ; Male ; Female ; Brain-Computer Interfaces ; Brain/physiology ; Wearable Electronic Devices ; Electrodes ; Young Adult ; },
abstract = {Electroencephalography (EEG) has emerged as a pivotal tool in both research and clinical practice due to its non-invasive nature, cost-effectiveness, and ability to provide real-time monitoring of brain activity. Wearable EEG technology opens new avenues for consumer applications, such as mental health monitoring, neurofeedback training, and brain-computer interfaces. However, there is still much to verify and re-examine regarding the functionality of these devices and the quality of the signal they capture, particularly as the field evolves rapidly. In this study, we recorded the resting-state brain activity of healthy volunteers via three consumer-grade EEG devices, namely PSBD Headband Pro, PSBD Headphones Lite, and Muse S Gen 2, and compared the spectral characteristics of the signal obtained with that recorded via the research-grade Brain Product amplifier (BP) with the mirroring montages. The results showed that all devices exhibited higher mean power in the low-frequency bands, which are characteristic of dry-electrode technology. PSBD Headband proved to match BP most precisely among the other examined devices. PSBD Headphones displayed a moderate correspondence with BP and signal quality issues in the central group of electrodes. Muse demonstrated the poorest signal quality, with extremely low alignment with BP. Overall, this study underscores the importance of considering device-specific design constraints and emphasizes the need for further validation to ensure the reliability and accuracy of wearable EEG devices.},
}

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

@article {pmid39771785,
year = {2024},
author = {Novičić, M and Djordjević, O and Miler-Jerković, V and Konstantinović, L and Savić, AM},
title = {Improving the Performance of Electrotactile Brain-Computer Interface Using Machine Learning Methods on Multi-Channel Features of Somatosensory Event-Related Potentials.},
journal = {Sensors (Basel, Switzerland)},
volume = {24},
number = {24},
pages = {},
pmid = {39771785},
issn = {1424-8220},
support = {6066223//Science Fund of the Republic of Serbia/ ; },
mesh = {Humans ; *Brain-Computer Interfaces ; *Electroencephalography/methods ; *Evoked Potentials, Somatosensory/physiology ; Male ; *Machine Learning ; Adult ; Female ; Algorithms ; Touch/physiology ; Young Adult ; Signal Processing, Computer-Assisted ; Support Vector Machine ; },
abstract = {Traditional tactile brain-computer interfaces (BCIs), particularly those based on steady-state somatosensory-evoked potentials, face challenges such as lower accuracy, reduced bit rates, and the need for spatially distant stimulation points. In contrast, using transient electrical stimuli offers a promising alternative for generating tactile BCI control signals: somatosensory event-related potentials (sERPs). This study aimed to optimize the performance of a novel electrotactile BCI by employing advanced feature extraction and machine learning techniques on sERP signals for the classification of users' selective tactile attention. The experimental protocol involved ten healthy subjects performing a tactile attention task, with EEG signals recorded from five EEG channels over the sensory-motor cortex. We employed sequential forward selection (SFS) of features from temporal sERP waveforms of all EEG channels. We systematically tested classification performance using machine learning algorithms, including logistic regression, k-nearest neighbors, support vector machines, random forests, and artificial neural networks. We explored the effects of the number of stimuli required to obtain sERP features for classification and their influence on accuracy and information transfer rate. Our approach indicated significant improvements in classification accuracy compared to previous studies. We demonstrated that the number of stimuli for sERP generation can be reduced while increasing the information transfer rate without a statistically significant decrease in classification accuracy. In the case of the support vector machine classifier, we achieved a mean accuracy over 90% for 10 electrical stimuli, while for 6 stimuli, the accuracy decreased by less than 7%, and the information transfer rate increased by 60%. This research advances methods for tactile BCI control based on event-related potentials. This work is significant since tactile stimulation is an understudied modality for BCI control, and electrically induced sERPs are the least studied control signals in reactive BCIs. Exploring and optimizing the parameters of sERP elicitation, as well as feature extraction and classification methods, is crucial for addressing the accuracy versus speed trade-off in various assistive BCI applications where the tactile modality may have added value.},
}

Humans
*Brain-Computer Interfaces
*Electroencephalography/methods
*Evoked Potentials, Somatosensory/physiology
Male
*Machine Learning
Adult
Female
Algorithms
Touch/physiology
Young Adult
Signal Processing, Computer-Assisted
Support Vector Machine

@article {pmid39771721,
year = {2024},
author = {Víg, L and Zátonyi, A and Csernyus, B and Horváth, ÁC and Bojtár, M and Kele, P and Madarász, M and Rózsa, B and Fürjes, P and Hermann, P and Hakkel, O and Péter, L and Fekete, Z},
title = {Optically Controlled Drug Delivery Through Microscale Brain-Machine Interfaces Using Integrated Upconverting Nanoparticles.},
journal = {Sensors (Basel, Switzerland)},
volume = {24},
number = {24},
pages = {},
pmid = {39771721},
issn = {1424-8220},
support = {TKP2021-EGA-42//National Research, Development and Innovation Office/ ; TKP2021-EGA-04//National Research, Development and Innovation Office/ ; VKE-2018-00032//National Research, Development and Innovation Office/ ; KFI-2018-00097//National Research, Development and Innovation Office/ ; 2020-2.1.1-ED-2022-00208//National Research, Development and Innovation Office/ ; (NAP2022I-8/2022//Hungarian Academy of Sciences/ ; Bolyai Janos Scholarship//Hungarian Academy of Sciences/ ; NKFIH FK 134403//National Research, Development and Innovation Office/ ; },
mesh = {*Nanoparticles/chemistry ; *Brain-Computer Interfaces ; *Drug Delivery Systems/methods/instrumentation ; Animals ; Brain/physiology ; Rats ; },
abstract = {The aim of this work is to incorporate lanthanide-cored upconversion nanoparticles (UCNP) into the surface of microengineered biomedical implants to create a spatially controlled and optically releasable model drug delivery device in an integrated fashion. Our approach enables silicone-based microelectrocorticography (ECoG) implants holding platinum/iridium recording sites to serve as a stable host of UCNPs. Nanoparticles excitable in the near-infrared (lower energy) regime and emitting visible (higher energy) light are utilized in a study. With the upconverted higher energy photons, we demonstrate the induction of photochemical (cleaving) reactions that enable the local release of specific dyes as a model system near the implant. The modified ECoG electrodes can be implanted in brain tissue to act as an uncaging system that releases small amounts of substance while simultaneously measuring the evoked neural response upon light activation. In this paper, several technological challenges like the surface modification of UCNPs, the immobilization of particles on the implantable platform, and measuring the stability of integrated UCNPs in in vitro and in vivo conditions are addressed in detail. Besides the chemical, mechanical, and optical characterization of the ready-to-use devices, the effect of nanoparticles on the original electrophysiological function is also evaluated. The results confirm that silicone-based brain-machine interfaces can be efficiently complemented with UCNPs to facilitate local model drug release.},
}

*Nanoparticles/chemistry
*Brain-Computer Interfaces
*Drug Delivery Systems/methods/instrumentation
Animals
Brain/physiology
Rats

@article {pmid39771689,
year = {2024},
author = {Fang, F and Gao, T and Wu, J},
title = {Humanity Test-EEG Data Mediated Artificial Intelligence Multi-Person Interactive System.},
journal = {Sensors (Basel, Switzerland)},
volume = {24},
number = {24},
pages = {},
pmid = {39771689},
issn = {1424-8220},
mesh = {Humans ; *Electroencephalography/methods ; *Artificial Intelligence ; *Emotions/physiology ; Male ; Female ; Adult ; Young Adult ; },
abstract = {Artificial intelligence (AI) systems are widely applied in various industries and everyday life, particularly in fields such as virtual assistants, healthcare, and education. However, this paper highlights that existing research has often overlooked the philosophical and media aspects. To address this, we developed an interactive system called "Human Nature Test". In this context, "human nature" refers to emotion and consciousness, while "test" involves a critical analysis of AI technology and an exploration of the differences between humanity and technicality. Additionally, through experimental research and literature analysis, we found that the integration of electroencephalogram (EEG) data with AI systems is becoming a significant trend. The experiment involved 20 participants, with two conditions: C1 (using EEG data) and C2 (without EEG data). The results indicated a significant increase in immersion under the C1 condition, along with a more positive emotional experience. We summarized three design directions: enhancing immersion, creating emotional experiences, and expressing philosophical concepts. Based on these findings, there is potential for further developing EEG data as a medium to enrich interactive experiences, offering new insights into the fusion of technology and human emotion.},
}

Humans
*Electroencephalography/methods
*Artificial Intelligence
*Emotions/physiology
Male
Female
Adult
Young Adult

@article {pmid39771656,
year = {2024},
author = {Khalil, AEK and Perez-Diaz, JA and Cantoral-Ceballos, JA and Antelis, JM},
title = {Unlocking Security for Comprehensive Electroencephalogram-Based User Authentication Systems.},
journal = {Sensors (Basel, Switzerland)},
volume = {24},
number = {24},
pages = {},
pmid = {39771656},
issn = {1424-8220},
support = {N/A//Tecnológico de Monterrey/ ; },
mesh = {*Electroencephalography/methods ; Humans ; *Neural Networks, Computer ; Computer Security ; Signal Processing, Computer-Assisted ; Algorithms ; Brain-Computer Interfaces ; Brain/physiology ; },
abstract = {With recent significant advancements in artificial intelligence, the necessity for more reliable recognition systems has rapidly increased to safeguard individual assets. The use of brain signals for authentication has gained substantial interest within the scientific community over the past decade. Most previous efforts have focused on identifying distinctive information within electroencephalogram (EEG) recordings. In this study, an EEG-based user authentication scheme is presented, employing a multi-layer perceptron feedforward neural network (MLP FFNN). The scheme utilizes P300 potentials derived from EEG signals, focusing on the user's intent to select specific characters. This approach involves two phases: user identification and user authentication. Both phases utilize EEG recordings of brain signals, data preprocessing, a database to store and manage these recordings for efficient retrieval and organization, and feature extraction using mutual information (MI) from selected EEG data segments, specifically targeting power spectral density (PSD) across five frequency bands. The user identification phase employs multi-class classifiers to predict the identity of a user from a set of enrolled users. The user authentication phase associates the predicted user identities with user labels using probability assessments, verifying the claimed identity as either genuine or an impostor. This scheme combines EEG data segments with user mapping, confidence calculations, and claimed user verification for robust authentication. It also accommodates new users by transforming EEG data into feature vectors without the need for retraining. The model extracts selected features to identify users and to classify the input based on these features to authenticate the user. The experiments show that the proposed scheme can achieve 97% accuracy in EEG-based user identification and authentication.},
}

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

@article {pmid39768099,
year = {2024},
author = {Mahmoud, TSM and Munawar, A and Nawaz, MZ and Chen, Y},
title = {Enhancing Multispectral Breast Imaging Quality Through Frame Accumulation and Hybrid GA-CPSO Registration.},
journal = {Bioengineering (Basel, Switzerland)},
volume = {11},
number = {12},
pages = {},
pmid = {39768099},
issn = {2306-5354},
abstract = {Multispectral transmission imaging has emerged as a promising technique for imaging breast tissue with high resolution. However, the method encounters challenges such as low grayscale, noisy transmission images with weak signals, primarily due to the strong absorption and scattering of light in breast tissue. A common approach to improve the signal-to-noise ratio (SNR) and overall image quality is frame accumulation. However, factors such as camera jitter and respiratory motion during image acquisition can cause frame misalignment, degrading the quality of the accumulated image. To address these issues, this study proposes a novel image registration method. A hybrid approach combining a genetic algorithm (GA) and a constriction factor-based particle swarm optimization (CPSO), referred to as GA-CPSO, is applied for image registration before frame accumulation. The efficiency of this hybrid method is enhanced by incorporating a squared constriction factor (SCF), which speeds up the registration process and improves convergence towards optimal solutions. The GA identifies potential solutions, which are then refined by CPSO to expedite convergence. This methodology was validated on the sequence of breast frames taken at 600 nm, 620 nm, 670 nm, and 760 nm wavelength of light and proved the enhancement of accuracy by various mathematical assessments. It demonstrated high accuracy (99.93%) and reduced registration time. As a result, the GA-CPSO approach significantly improves the effectiveness of frame accumulation and enhances overall image quality. This study explored the groundwork for precise multispectral transmission image segmentation and classification.},
}

@article {pmid39768034,
year = {2024},
author = {Geravanchizadeh, M and Shaygan Asl, A and Danishvar, S},
title = {Selective Auditory Attention Detection Using Combined Transformer and Convolutional Graph Neural Networks.},
journal = {Bioengineering (Basel, Switzerland)},
volume = {11},
number = {12},
pages = {},
pmid = {39768034},
issn = {2306-5354},
abstract = {Attention is one of many human cognitive functions that are essential in everyday life. Given our limited processing capacity, attention helps us focus only on what matters. Focusing attention on one speaker in an environment with many speakers is a critical ability of the human auditory system. This paper proposes a new end-to-end method based on the combined transformer and graph convolutional neural network (TraGCNN) that can effectively detect auditory attention from electroencephalograms (EEGs). This approach eliminates the need for manual feature extraction, which is often time-consuming and subjective. Here, the first EEG signals are converted to graphs. We then extract attention information from these graphs using spatial and temporal approaches. Finally, our models are trained with these data. Our model can detect auditory attention in both the spatial and temporal domains. Here, the EEG input is first processed by transformer layers to obtain a sequential representation of EEG based on attention onsets. Then, a family of graph convolutional layers is used to find the most active electrodes using the spatial position of electrodes. Finally, the corresponding EEG features of active electrodes are fed into the graph attention layers to detect auditory attention. The Fuglsang 2020 dataset is used in the experiments to train and test the proposed and baseline systems. The new TraGCNN approach, as compared with state-of-the-art attention classification methods from the literature, yields the highest performance in terms of accuracy (80.12%) as a classification metric. Additionally, the proposed model results in higher performance than our previously graph-based model for different lengths of EEG segments. The new TraGCNN approach is advantageous because attenuation detection is achieved from EEG signals of subjects without requiring speech stimuli, as is the case with conventional auditory attention detection methods. Furthermore, examining the proposed model for different lengths of EEG segments shows that the model is faster than our previous graph-based detection method in terms of computational complexity. The findings of this study have important implications for the understanding and assessment of auditory attention, which is crucial for many applications, such as brain-computer interface (BCI) systems, speech separation, and neuro-steered hearing aid development.},
}

@article {pmid39766488,
year = {2024},
author = {Ma, Y and Huang, Z and Yang, Y and Zhang, S and Dong, Q and Wang, R and Hu, L},
title = {Emotion Recognition Model of EEG Signals Based on Double Attention Mechanism.},
journal = {Brain sciences},
volume = {14},
number = {12},
pages = {},
pmid = {39766488},
issn = {2076-3425},
abstract = {BACKGROUND: Emotions play a crucial role in people's lives, profoundly affecting their cognition, decision-making, and interpersonal communication. Emotion recognition based on brain signals has become a significant challenge in the fields of affective computing and human-computer interaction.

METHODS: Addressing the issue of inaccurate feature extraction and low accuracy of existing deep learning models in emotion recognition, this paper proposes a multi-channel automatic classification model for emotion EEG signals named DACB, which is based on dual attention mechanisms, convolutional neural networks, and bidirectional long short-term memory networks. DACB extracts features in both temporal and spatial dimensions, incorporating not only convolutional neural networks but also SE attention mechanism modules for learning the importance of different channel features, thereby enhancing the network's performance. DACB also introduces dot product attention mechanisms to learn the importance of spatial and temporal features, effectively improving the model's accuracy.

RESULTS: The accuracy of this method in single-shot validation tests on the SEED-IV and DREAMER (Valence-Arousal-Dominance three-classification) datasets is 99.96% and 87.52%, 90.06%, and 89.05%, respectively. In 10-fold cross-validation tests, the accuracy is 99.73% and 84.26%, 85.40%, and 85.02%, outperforming other models.

CONCLUSIONS: This demonstrates that the DACB model achieves high accuracy in emotion classification tasks, demonstrating outstanding performance and generalization ability and providing new directions for future research in EEG signal recognition.},
}

@article {pmid39766471,
year = {2024},
author = {Adolf, A and Köllőd, CM and Márton, G and Fadel, W and Ulbert, I},
title = {The Effect of Processing Techniques on the Classification Accuracy of Brain-Computer Interface Systems.},
journal = {Brain sciences},
volume = {14},
number = {12},
pages = {},
pmid = {39766471},
issn = {2076-3425},
support = {KDP-2021-12, 1022428/001//National Research, Development and Innovation Office/ ; FK146115//National Research, Development and Innovation Office/ ; NAP2022-I-2/2022//Hungarian Academy of Sciences/ ; RRF-2.3.1-21-2022-00015//National Research, Development and Innovation Office/ ; },
abstract = {Background/Objectives: Accurately classifying Electroencephalography (EEG) signals is essential for the effective operation of Brain-Computer Interfaces (BCI), which is needed for reliable neurorehabilitation applications. However, many factors in the processing pipeline can influence classification performance. The objective of this study is to assess the effects of different processing steps on classification accuracy in EEG-based BCI systems. Methods: This study explores the impact of various processing techniques and stages, including the FASTER algorithm for artifact rejection (AR), frequency filtering, transfer learning, and cropped training. The Physionet dataset, consisting of four motor imagery classes, was used as input due to its relatively large number of subjects. The raw EEG was tested with EEGNet and Shallow ConvNet. To examine the impact of adding a spatial dimension to the input data, we also used the Multi-branch Conv3D Net and developed two new models, Conv2D Net and Conv3D Net. Results: Our analysis showed that classification accuracy can be affected by many factors at every stage. Applying the AR method, for instance, can either enhance or degrade classification performance, depending on the subject and the specific network architecture. Transfer learning was effective in improving the performance of all networks for both raw and artifact-rejected data. However, the improvement in classification accuracy for artifact-rejected data was less pronounced compared to unfiltered data, resulting in reduced precision. For instance, the best classifier achieved 46.1% accuracy on unfiltered data, which increased to 63.5% with transfer learning. In the filtered case, accuracy rose from 45.5% to only 55.9% when transfer learning was applied. An unexpected outcome regarding frequency filtering was observed: networks demonstrated better classification performance when focusing on lower-frequency components. Higher frequency ranges were more discriminative for EEGNet and Shallow ConvNet, but only when cropped training was applied. Conclusions: The findings of this study highlight the complex interaction between processing techniques and neural network performance, emphasizing the necessity for customized processing approaches tailored to specific subjects and network architectures.},
}

@article {pmid39766454,
year = {2024},
author = {Beck, S and Liberman, Y and Dubljević, V},
title = {Media Representation of the Ethical Issues Pertaining to Brain-Computer Interface (BCI) Technology.},
journal = {Brain sciences},
volume = {14},
number = {12},
pages = {},
pmid = {39766454},
issn = {2076-3425},
abstract = {BACKGROUND/OBJECTIVES: Brain-computer interfaces (BCIs) are a rapidly developing technology that captures and transmits brain signals to external sources, allowing the user control of devices such as prosthetics. BCI technology offers the potential to restore physical capabilities in the body and change how we interact and communicate with computers and each other. While BCI technology has existed for decades, recent developments have caused the technology to generate a host of ethical issues and discussions in both academic and public circles. Given that media representation has the potential to shape public perception and policy, it is necessary to evaluate the space that these issues take in public discourse.

METHODS: We conducted a rapid review of media articles in English discussing ethical issues of BCI technology from 2013 to 2024 as indexed by LexisNexis. Our searches yielded 675 articles, with a final sample containing 182 articles. We assessed the themes of the articles and coded them based on the ethical issues discussed, ethical frameworks, recommendations, tone, and application of technology.

RESULTS: Our results showed a marked rise in interest in media articles over time, signaling an increased focus on this topic. The majority of articles adopted a balanced or neutral tone when discussing BCIs and focused on ethical issues regarding privacy, autonomy, and regulation.

CONCLUSIONS: Current discussion of ethical issues reflects growing news coverage of companies such as Neuralink, and reveals a mounting distrust of BCI technology. The growing recognition of ethical considerations in BCI highlights the importance of ethical discourse in shaping the future of the field.},
}

@article {pmid39760490,
year = {2025},
author = {Zhai, H and Li, P and Wang, H and Wang, X},
title = {DMSO-promoted α-bromination of α-aryl ketones for the construction of 2-aryl-2-bromo-cycloketones.},
journal = {Organic & biomolecular chemistry},
volume = {},
number = {},
pages = {},
doi = {10.1039/d4ob01937g},
pmid = {39760490},
issn = {1477-0539},
abstract = {A DMSO-promoted practical one-step α-bromination reaction of α-aryl ketones with NBS has been developed for the construction of 2-aryl-2-bromo-cycloketones. The desired regioselective α-bromination products were isolated in moderate to good yields, with a maximum tested scale of 15 mmol. Notably, ketamine derivatives could be smoothly synthesized in two steps.},
}

@article {pmid39760422,
year = {2025},
author = {Zhu, L and Wang, Y and Huang, A and Tan, X and Zhang, J},
title = {A multi-branch, multi-scale, and multi-view CNN with lightweight temporal attention mechanism for EEG-based motor imagery decoding.},
journal = {Computer methods in biomechanics and biomedical engineering},
volume = {},
number = {},
pages = {1-15},
doi = {10.1080/10255842.2024.2448576},
pmid = {39760422},
issn = {1476-8259},
abstract = {Convolutional neural networks (CNNs) have been widely utilized for decoding motor imagery (MI) from electroencephalogram (EEG) signals. However, extracting discriminative spatial-temporal-spectral features from low signal-to-noise ratio EEG signals remains challenging. This paper proposes MBMSNet , a multi-branch, multi-scale, and multi-view CNN with a lightweight temporal attention mechanism for EEG-Based MI decoding. Specifically, MBMSNet first extracts multi-view representations from raw EEG signals, followed by independent branches to capture spatial, spectral, temporal-spatial, and temporal-spectral features. Each branch includes a domain-specific convolutional layer, a variance layer, and a temporal attention layer. Finally, the features derived from each branch are concatenated with weights and classified through a fully connected layer. Experiments demonstrate MBMSNet outperforms state-of-the-art models, achieving accuracies of 84.60% on BCI Competition IV 2a, 87.80% on 2b, and 74.58% on OpenBMI, showcasing its potential for robust BCI applications.},
}

@article {pmid39759760,
year = {2024},
author = {Cisotto, G and Zancanaro, A and Zoppis, IF and Manzoni, SL},
title = {hvEEGNet: a novel deep learning model for high-fidelity EEG reconstruction.},
journal = {Frontiers in neuroinformatics},
volume = {18},
number = {},
pages = {1459970},
pmid = {39759760},
issn = {1662-5196},
abstract = {INTRODUCTION: Modeling multi-channel electroencephalographic (EEG) time-series is a challenging tasks, even for the most recent deep learning approaches. Particularly, in this work, we targeted our efforts to the high-fidelity reconstruction of this type of data, as this is of key relevance for several applications such as classification, anomaly detection, automatic labeling, and brain-computer interfaces.

METHODS: We analyzed the most recent works finding that high-fidelity reconstruction is seriously challenged by the complex dynamics of the EEG signals and the large inter-subject variability. So far, previous works provided good results in either high-fidelity reconstruction of single-channel signals, or poor-quality reconstruction of multi-channel datasets. Therefore, in this paper, we present a novel deep learning model, called hvEEGNet, designed as a hierarchical variational autoencoder and trained with a new loss function. We tested it on the benchmark Dataset 2a (including 22-channel EEG data from 9 subjects).

RESULTS: We show that it is able to reconstruct all EEG channels with high-fidelity, fastly (in a few tens of epochs), and with high consistency across different subjects. We also investigated the relationship between reconstruction fidelity and the training duration and, using hvEEGNet as an anomaly detector, we spotted some data in the benchmark dataset that are corrupted and never highlighted before.

DISCUSSION: Thus, hvEEGNet could be very useful in several applications where automatic labeling of large EEG dataset is needed and time-consuming. At the same time, this work opens new fundamental research questions about (1) the effectiveness of deep learning models training (for EEG data) and (2) the need for a systematic characterization of the input EEG data to ensure robust modeling.},
}

@article {pmid39759080,
year = {2024},
author = {Zhao, X and Xu, S and Geng, K and Zhou, T and Xu, T and Wang, Z and Feng, S and Hu, H},
title = {MP: A steady-state visual evoked potential dataset based on multiple paradigms.},
journal = {iScience},
volume = {27},
number = {11},
pages = {111030},
pmid = {39759080},
issn = {2589-0042},
abstract = {In the field of steady-state visual evoked potential (SSVEP), stimulus paradigms are regularly arranged or mimic the style of a keyboard with the same size. However, stimulation paradigms have important effects on the performance of SSVEP systems, which correlate with the electroencephalogram (EEG) signal amplitude and recognition accuracy. This paper provides MP dataset that was acquired using a 12-target BCI speller. MP dataset contains 9-channel EEG signals from the occipital region of 24 subjects under 5 stimulation paradigms with different stimulus sizes and arrangements. Stimuli were encoded using joint frequency and phase modulation (JFPM) method. Subjects completed an offline prompted spelling task using a speller under 5 paradigms. Each experiment contains 8 blocks, and each block contains 12 trials. Designers can use this dataset to test the performance of algorithms considering "stimulus size" and "stimulus arrangement". EEG data showed SSVEP features through amplitude-frequency analysis. FBCCA and TRCA confirmed its suitability.},
}

@article {pmid39758181,
year = {2025},
author = {Ammar, A and Salem, A and Simak, ML and Horst, F and Schöllhorn, WI},
title = {Acute effects of motor learning models on technical efficiency in strength-coordination exercises: a comparative analysis of Olympic snatch biomechanics in beginners.},
journal = {Biology of sport},
volume = {42},
number = {1},
pages = {151-161},
pmid = {39758181},
issn = {0860-021X},
abstract = {Despite the development of various motor learning models over many decades, the question of which model is most effective under which conditions to optimize the acquisition of skills remains a heated and recurring debate. This is particularly important in connection with learning sports movements with a high strength component. This study aims to examine the acute effects of various motor learning models on technical efficiency and force production during the Olympic snatch movement. In a within-subject design, sixteen highly active male participants (mean age: 23.13 ± 2.09 years), who were absolute beginners regarding the learning task, engaged in randomized snatch learning bouts, consisting of 36 trials across different learning models: differential learning (DL), contextual interference (serial, sCI; and blocked, bCI), and repetitive learning (RL). Kinematic and kinetic data were collected from three snatch trials executed following each learning bout. Discrete data from the most commonly monitored biomechanical parameters in Olympic weightlifting were analyzed using inferential statistics to identify differences between learning models. The statistical analysis revealed no significant differences between the learning models across all tested parameters, with p-values ranging from 0.236 to 0.99. However, it was observed that only the bouts with an exercise sequence following the DL model resulted in an average antero-posterior displacement of the barbell that matched the optimal displacement. This was characterized by a mean positive displacement towards the lifter during the pulling phases, a negative displacement away from the lifter in the turnover phase, and a return to positive displacement in the catch phase. These findings indicate the limited acute impact of the exercise sequences based on the three motor learning models on Olympic snatch technical efficiency in beginners, yet they hint at a possible slight advantage for the DL model. Coaches might therefore consider incorporating the DL model to potentially enhance technical efficiency, especially during the early stages of skill acquisition. Future research, involving even bigger amounts of exercise noise, longer learning periods, or a greater number of total learning trials and sessions, is essential to verify the potential advantages of the DL model for weightlifting technical efficiency.},
}

@article {pmid39758128,
year = {2025},
author = {Wang, B and Zhang, Y and Li, H and Dou, H and Guo, Y and Deng, Y},
title = {Biologically inspired heterogeneous learning for accurate, efficient and low-latency neural network.},
journal = {National science review},
volume = {12},
number = {1},
pages = {nwae301},
pmid = {39758128},
issn = {2053-714X},
abstract = {The pursuit of artificial neural networks that mirror the accuracy, efficiency and low latency of biological neural networks remains a cornerstone of artificial intelligence (AI) research. Here, we incorporated recent neuroscientific findings of self-inhibiting autapse and neuron heterogeneity for innovating a spiking neural network (SNN) with enhanced learning and memorizing capacities. A bi-level programming paradigm was formulated to respectively learn neuron-level biophysical variables and network-level synapse weights for nested heterogeneous learning. We successfully demonstrated that our biologically inspired neuron model could reproduce neural statistics at both individual and group levels, contributing to the effective decoding of brain-computer interface data. Furthermore, the heterogeneous SNN showed higher accuracy (1%-10% improvement), superior efficiency (maximal 17.83-fold reduction in energy) and lower latency (maximal 5-fold improvement) in performing several AI tasks. For the first time, we benchmarked SNN for conducting cell type identification from scRNA-seq data. The proposed model correctly identified very rare cell types associated with severe brain diseases where typical SNNs failed.},
}

@article {pmid39757218,
year = {2025},
author = {Kim, MS and Park, H and Kwon, I and An, KO and Kim, H and Park, G and Hyung, W and Im, CH and Shin, JH},
title = {Efficacy of brain-computer interface training with motor imagery-contingent feedback in improving upper limb function and neuroplasticity among persons with chronic stroke: a double-blinded, parallel-group, randomized controlled trial.},
journal = {Journal of neuroengineering and rehabilitation},
volume = {22},
number = {1},
pages = {1},
pmid = {39757218},
issn = {1743-0003},
support = {NRCTR-IN20001//Translational Research Program for Rehabilitation Robots, National Rehabilitation Center, Ministry of Health and Welfare, Republic of Korea/ ; },
mesh = {Humans ; *Brain-Computer Interfaces ; *Neuronal Plasticity/physiology ; Male ; Female ; *Stroke Rehabilitation/methods ; Middle Aged ; *Upper Extremity/physiopathology ; Double-Blind Method ; Stroke/physiopathology ; Aged ; Electroencephalography ; Recovery of Function/physiology ; Adult ; Imagery, Psychotherapy/methods ; Chronic Disease ; Treatment Outcome ; Imagination/physiology ; },
abstract = {BACKGROUND: Brain-computer interface (BCI) technology can enhance neural plasticity and motor recovery in persons with stroke. However, the effects of BCI training with motor imagery (MI)-contingent feedback versus MI-independent feedback remain unclear. This study aimed to investigate whether the contingent connection between MI-induced brain activity and feedback influences functional and neural plasticity outcomes. We hypothesized that BCI training, with MI-contingent feedback, would result in greater improvements in upper limb function and neural plasticity compared to BCI training, with MI-independent feedback.

METHODS: This randomized controlled trial included persons with chronic stroke who underwent BCI training involving functional electrical stimulation feedback on the affected wrist extensor. Primary outcomes included the Medical Research Council (MRC) scale score for muscle strength in the wrist extensor (MRC-WE) and active range of motion in wrist extension (AROM-WE). Resting-state electroencephalogram recordings were used to assess neural plasticity.

RESULTS: Compared to the MI-independent feedback BCI group, the MI-contingent feedback BCI group showed significantly greater improvements in MRC-WE scores (mean difference = 0.52, 95% CI = 0.03-1.00, p = 0.036) and demonstrated increased AROM-WE at 4 weeks post-intervention (p = 0.019). Enhanced functional connectivity in the affected hemisphere was observed in the MI-contingent feedback BCI group, correlating with MRC-WE and Fugl-Meyer assessment-distal scores. Improvements were also observed in the unaffected hemisphere's functional connectivity.

CONCLUSIONS: BCI training with MI-contingent feedback is more effective than MI-independent feedback in improving AROM-WE, MRC, and neural plasticity in individuals with chronic stroke. BCI technology could be a valuable addition to conventional rehabilitation for stroke survivors, enhancing recovery outcomes.

TRIAL REGISTRATION: CRIS (KCT0009013).},
}

Humans
*Brain-Computer Interfaces
*Neuronal Plasticity/physiology
Male
Female
*Stroke Rehabilitation/methods
Middle Aged
*Upper Extremity/physiopathology
Double-Blind Method
Stroke/physiopathology
Aged
Electroencephalography
Recovery of Function/physiology
Adult
Imagery, Psychotherapy/methods
Chronic Disease
Treatment Outcome
Imagination/physiology

@article {pmid39755291,
year = {2025},
author = {Vogeley, AO and Livinski, AA and Varnosfaderani, SD and Javaheripour, N and Jamalabadi, H and Kotoula, V and Henter, ID and Hejazi, NS and Price, RB and Yavi, M and Walter, M and Zarate, CA and Kheirkhah, M},
title = {Temporal Dynamics of Affective Scene Processing in the Healthy Adult Human Brain.},
journal = {Neuroscience and biobehavioral reviews},
volume = {},
number = {},
pages = {106003},
doi = {10.1016/j.neubiorev.2025.106003},
pmid = {39755291},
issn = {1873-7528},
abstract = {Understanding how the brain distinguishes emotional from neutral scenes is crucial for advancing brain-computer interfaces, enabling real-time emotion detection for faster, more effective responses, and improving treatments for emotional disorders like depression and anxiety. However, inconsistent research findings have arisen from differences in study settings, such as variations in the time windows, brain regions, and emotion categories examined across studies. This review sought to compile the existing literature on the timing at which the adult brain differentiates basic affective from neutral scenes in less than one second, as previous studies have consistently shown that the brain can begin recognizing emotions within just a few milliseconds. The review includes studies that used electroencephalography (EEG) or magnetoencephalography (MEG) in healthy adults to examine brain responses to emotional versus neutral images within one second. Articles of interest were limited to the English language but not to any publication year. Excluded studies involved only patients (of any diagnosis), participants under age 18 (since emotional processing can differ between adults and younger individuals), non-passive tasks, low temporal resolution techniques, time intervals over one second, and animals. Of the 3,045 screened articles, 19 met these criteria. Despite the variations between studies, the earliest onset for heightened brain responses to basic affective scenes compared to neutral ones was most commonly observed within the 250-300 ms time window. To the best of our knowledge, this review is the first to synthesize data on the timing of brain differentiation between emotional and neutral scenes in healthy adults.},
}

@article {pmid39755229,
year = {2025},
author = {Ma, X and Xue, S and Ma, H and Saeed, S and Zhang, Y and Meng, Y and Chen, H and Yu, H and Wang, H and Hu, S and Cai, M},
title = {Esketamine alleviates LPS-induced depression-like behavior by activating Nrf2-mediated anti-inflammatory response in adolescent mice.},
journal = {Neuroscience},
volume = {},
number = {},
pages = {},
doi = {10.1016/j.neuroscience.2024.12.062},
pmid = {39755229},
issn = {1873-7544},
abstract = {BACKGROUND: The mechanisms underlying esketamine's therapeutic effects remain elusive. The study aimed to explore the impact of single esketamine treatment on LPS-induced adolescent depressive-like behaviors and the role of Nrf2 regulated neuroinflammatory response in esketamine-produced rapid antidepressant efficacy.

METHODS: Adolescent male C57BL/6J mice were randomly assigned to three groups: control, LPS, and LPS + esketamine (15 mg/kg, i.p.). Depressive-like behaviors were evaluated via the OFT, NFST, and TST. Protein expression of Nrf2 and inflammatory cytokines, including TNF-α, IL-1β, and iNOS in the hippocampus and mPFC, were measured by western blot. Moreover, the Nrf2 inhibitor, ML385, was also applied in the current study. The depressive-like behaviors and the protein expression of Nrf2, TNF-α, IL-1β, and iNOS in mPFC and hippocampus were also measured. Additionally, the plasma's pro-inflammatory cytokines and anti-inflammatory cytokines were assessed using ELISA methods with or without ML385.

RESULTS: A single administration of esketamine treatment alleviated the LPS-induced depressive-like behaviors. Esketamine increased the expression of Nrf2 and reduced the expression of the inflammatory cytokines, including TNF-α, IL-1β, and iNOS, in the mPFC and hippocampus. Notably, pharmacological inhibition of Nrf2 via ML385 administration abrogated the antidepressive-like behaviors and anti-inflammatory effects induced by esketamine. In the periphery, esketamine mitigated the LPS-induced elevation of pro-inflammatory cytokines, and the reduction of anti-inflammatory cytokines, and this effect was reversed by Nrf2 inhibition.

CONCLUSION: Esketamine treatment exerts rapid antidepressant effects and attenuates neuroinflammation in LPS-induced adolescent depressive-like behaviors, potentially through the activation of Nrf2-mediated anti-inflammatory signaling.},
}

@article {pmid39755222,
year = {2025},
author = {Huang, J and Wei, S and Gao, Z and Jiang, S and Wang, M and Sun, L and Ding, W and Zhang, D},
title = {Local structural-functional coupling with counterfactual explanations for epilepsy prediction.},
journal = {NeuroImage},
volume = {306},
number = {},
pages = {120978},
doi = {10.1016/j.neuroimage.2024.120978},
pmid = {39755222},
issn = {1095-9572},
abstract = {The structural-functional brain connections coupling (SC-FC coupling) describes the relationship between white matter structural connections (SC) and the corresponding functional activation or functional connections (FC). It has been widely used to identify brain disorders. However, the existing research on SC-FC coupling focuses on global and regional scales, and few studies have investigated the impact of brain disorders on this relationship from the perspective of multi-brain region cooperation (i.e., local scale). Here, we propose the local SC-FC coupling pattern for brain disorders prediction. Compared with previous methods, the proposed patterns quantify the relationship between SC and FC in terms of subgraphs rather than whole connections or single brain regions. Specifically, we first construct structural and functional connections using diffusion tensor imaging (DTI) and resting-state functional magnetic resonance imaging (rs-fMRI) data, subsequently organizing them into a multimodal brain network. Then, we extract subgraphs from these multimodal brain networks and select them based on their frequencies to generate local SC-FC coupling patterns. Finally, we employ these patterns to identify brain disorders while refining abnormal patterns to generate counterfactual explanations. Results on a real epilepsy dataset suggest that the proposed method not only outperforms existing methods in accuracy but also provides insights into the local SC-FC coupling pattern and their changes in brain disorders. Code available at https://github.com/UAIBC-Brain/Local-SC-FC-coupling-pattern.},
}

@article {pmid39755127,
year = {2025},
author = {Chu, T and Si, X and Song, X and Che, K and Dong, F and Guo, Y and Chen, D and Yao, W and Zhao, F and Xie, H and Shi, Y and Ma, H and Ming, D and Mao, N},
title = {Understanding structural-functional connectivity coupling in patients with major depressive disorder: A white matter perspective.},
journal = {Journal of affective disorders},
volume = {373},
number = {},
pages = {219-226},
doi = {10.1016/j.jad.2024.12.082},
pmid = {39755127},
issn = {1573-2517},
abstract = {PURPOSE: To elucidate the structural-functional connectivity (SC-FC) coupling in white matter (WM) tracts in patients with major depressive disorder (MDD).

METHODS: A total of 178 individuals diagnosed with MDD and 173 healthy controls (HCs) were recruited for this study. The Euclidean distance was calculated to assess SC-FC coupling. The primary analyses focused on investigating alterations in SC-FC coupling in WM tracts of individuals with MDD. Additionally, we explored the association between coupling and clinical symptoms. Secondary analyses examined differences among three subgroups of MDD: those with suicidal ideation (SI), those with a history of suicidal attempts (SA), and those non-suicidal (NS).

RESULTS: The study revealed increased SC-FC coupling mainly in the middle cerebellar peduncle and bilateral corticospinal tract (PFDR < 0.05) in patients with MDD compared with HCs. Additionally, right cerebral peduncle coupling strength exhibited a significant positive correlation with Hamilton Anxiety Scale scores (r = 0.269, PFDR = 0.041), while right cingulum (hippocampus) coupling strength showed a significant negative correlation with Nurses' Global Assessment of Suicide Risk scores (r = -0.159, PFDR = 0.036). An increase in left anterior limb of internal capsule (PBonferroni < 0.01) and left corticospinal tract (PBonferroni < 0.05) coupling has been observed in MDD with SI. Additionally, a decrease in right posterior limb of internal capsule coupling has been found in MDD with SA (PBonferroni < 0.05).

CONCLUSIONS: This study emphasizes the variations in SC-FC coupling in WM tracts in individuals with MDD and its subgroups, highlighting the crucial role of WM networks in the pathophysiology of MDD.},
}

@article {pmid39754303,
year = {2025},
author = {Qiu, C and Zhang, D and Wang, M and Mei, X and Chen, W and Yu, H and Yin, W and Peng, G and Hu, S},
title = {Peripheral Single-Cell Immune Characteristics Contribute to the Diagnosis of Alzheimer's Disease and Dementia With Lewy Bodies.},
journal = {CNS neuroscience & therapeutics},
volume = {31},
number = {1},
pages = {e70204},
pmid = {39754303},
issn = {1755-5949},
support = {2022YFC3602604//National Key Research and Development Program of China/ ; 2023YFC2506200//National Key Research and Development Program of China/ ; 2021R52016//Leading Talent of Scientific and Technological Innovation Ten Thousand Talents Program of Zhejiang Province/ ; 2020R01001//Innovation team for precision diagnosis and treatment of major brain diseases/ ; JNL 2023001B//Research Project of Jinan Microecological Biomedicine Shandong Laboratory/ ; 2021C03107//Zhejiang Provincial Key Research and Development Program/ ; 2022KTZ004//Chinese Medical Education Association/ ; 2022 F28//NINGBO Medical Health Leading Academic Discipline Project/ ; },
mesh = {Humans ; *Alzheimer Disease/immunology/diagnosis ; *Lewy Body Disease/immunology/diagnosis ; Aged ; Female ; Male ; *Leukocytes, Mononuclear/immunology/metabolism ; Aged, 80 and over ; Single-Cell Analysis/methods ; Middle Aged ; Biomarkers/blood ; T-Lymphocytes/immunology ; },
abstract = {OBJECTIVE: Alzheimer's disease (AD) and dementia with Lewy bodies (DLB) are common neurodegenerative diseases with distinct but overlapping pathogenic mechanisms. The clinical similarities between these diseases often result in high misdiagnosis rates, leading to serious consequences. Peripheral blood mononuclear cells (PBMCs) are easy to collect and can accurately reflect the immune characteristics of both DLB and AD.

METHODS: We utilized time-of-flight mass cytometry (CyTOF) with single-cell resolution to quantitatively analyze peripheral PBMCs, identifying 1228 immune characteristics. Based on the top-selected immune features, we constructed immunological elastic net (iEN) models.

RESULTS: These models demonstrated high diagnostic efficacy in distinguishing diseased samples from healthy donors as well as distinguishing AD and DLB cases. The selected features reveal that the primary peripheral immune characteristic of AD is a decrease in total T cells, while DLB is characterized by low expression of I-kappa-B-alpha (IKBα) in the classical monocyte subset.

CONCLUSIONS: These findings suggest that peripheral immune characteristics could serve as potential biomarkers, facilitating the diagnosis of neurodegenerative diseases.},
}

Humans
*Alzheimer Disease/immunology/diagnosis
*Lewy Body Disease/immunology/diagnosis
Aged
Female
Male
*Leukocytes, Mononuclear/immunology/metabolism
Aged, 80 and over
Single-Cell Analysis/methods
Middle Aged
Biomarkers/blood
T-Lymphocytes/immunology

@article {pmid39753090,
year = {2024},
author = {Pang, Y and Zhao, S and Zhang, Z and Xu, J and Gao, L and Zhang, R and Li, Z and Lu, F and Chen, H and Wu, H and Chen, M and Chen, K and Wang, J},
title = {Individual structural covariance connectome reveals aberrant brain developmental trajectories associated with childhood maltreatment.},
journal = {Journal of psychiatric research},
volume = {181},
number = {},
pages = {709-715},
doi = {10.1016/j.jpsychires.2024.12.032},
pmid = {39753090},
issn = {1879-1379},
abstract = {BACKGROUND: The long-term impact of childhood maltreatment (CM) on an individual's physical and mental health is suggested to be mediated by altered neurodevelopment. However, the exact neurobiological consequences of CM remain unclear.

METHODS: The present study aimed to investigate the relationship between CM and brain age based on structural magnetic resonance imaging data from a sample of 214 adults. The participants were divided into CM and non_CM groups according to Childhood Trauma Questionnaire. For each participant, brain connectome age was estimated from a large-scale structural covariance network through relevance vector regression. Brain predicted age difference (brain_PAD) was then calculated for each participant.

RESULTS: The brain connectome age matched well with chronological age in young adults (age range: 18-25 years) and adults (age range: 26-44 years) without CM, but not in individuals with CM. Compared with non_CM group, CM group was characterized by higher brain_PAD scores in young adults, whereas lower brain_PAD scores in adults. The finding revealed that brain development in individuals with CM seems to be accelerated in younger adults but retardation with increasing age. Moreover, individuals who suffered child abuse (but not neglect) showed higher brain_PAD scores than non_CM group, suggesting the different influence of abuse and neglect on neurodevelopment. Finally, the brain_PAD was positively correlated with attentional impulsivity in CM group.

CONCLUSIONS: CM affects different stages of adult brain development differently, and abuse and neglect have different influenced patterns, which may provide new evidence for the impact of CM on structural brain development.},
}

@article {pmid39752412,
year = {2025},
author = {Ding, C and Kim Geok, S and Sun, H and Roslan, S and Cao, S and Zhao, Y},
title = {Does music counteract mental fatigue? A systematic review.},
journal = {PloS one},
volume = {20},
number = {1},
pages = {e0316252},
pmid = {39752412},
issn = {1932-6203},
mesh = {*Mental Fatigue/prevention & control ; Humans ; *Music/psychology ; *Cognition/physiology ; Music Therapy/methods ; },
abstract = {INTRODUCTION: Mental fatigue, a psychobiological state induced by prolonged and sustained cognitive tasks, impairs both cognitive and physical performance. Several studies have investigated strategies to counteract mental fatigue. However, potential health risks and contextual restrictions often limit these strategies, which hinder their practical application. Due to its noninvasive and portable nature, music has been proposed as a promising strategy to counteract mental fatigue. However, the effects of music on performance decrements vary with different music styles. Synthesizing studies that systematically report music style and its impact on counteracting performance decrements is crucial for theoretical and practical applications.

OBJECTIVES: This review aims to provide a comprehensive systematic analysis of different music styles in counteracting mental fatigue and their effects on performance decrements induced by mental fatigue. Additionally, the mechanisms by which music counteracts mental fatigue will be discussed.

METHODS: A comprehensive search was conducted across five databases-Web of Science, PubMed, SCOPUS, SPORTDiscus via EBSCOhost, and the Psychological and Behavioral Sciences Collection via EBSCOhost-up to November 18, 2023. The selected studies focused solely on music interventions, with outcomes including subjective feelings of mental fatigue, physiological markers, and both cognitive and behavioral performance.

RESULTS: Nine studies met the predetermined criteria for inclusion in this review. The types of music interventions that counteract mental fatigue include relaxing, exciting, and personal preference music, all of which were associated with decreased subjective feelings of mental fatigue and changes in objective physiological markers. Cognitive performance, particularly in inhibition and working memory tasks impaired by mental fatigue, was countered by both relaxing and exciting music. Exciting music was found to decrease reaction time more effectively than relaxing music in working memory tasks. The physiological marker of steady-state visually evoked potential-based brain-computer interface (SSVEP-BCI) amplitude increased, confirming that exciting music counteracts mental fatigue more effectively than relaxing music. Behavioral performance in tasks such as arm-pointing, the Yo-Yo intermittent test, and the 5 km time-trial, which were impaired by mental fatigue, were counteracted by personal preference music.

CONCLUSION: Relaxing music, exciting music, and personal preference music effectively counteract mental fatigue by reducing feelings of fatigue and mitigating performance decrements. Individuals engaged in mentally demanding tasks can effectively counteract concurrent or subsequent cognitive performance decrements by simultaneously listening to relaxing or exciting music without lyrics or by using music during recovery from mental fatigue. Exciting music is more effective than relaxing music in counteracting mental fatigue. Personal preference music is effective in counteracting behavioral performance decrements in motor control and endurance tasks. Mentally fatigued individuals could apply personal preference music to counteract subsequent motor control performance decrements or simultaneously listen to it to counteract endurance performance decrements. Future studies should specify and examine the effects of different music genres, tempos, and intensities in counteracting mental fatigue. Additionally, the role of music in counteracting mental fatigue in contexts such as work productivity, traffic accident risk, and sports requires further investigation, along with the underlying mechanisms.},
}

*Mental Fatigue/prevention & control
Humans
*Music/psychology
*Cognition/physiology
Music Therapy/methods

@article {pmid39752045,
year = {2025},
author = {Xin, Q and Hu, H},
title = {To Attack or Not: A Neural Circuit Coding Sexually Dimorphic Aggression.},
journal = {Neuroscience bulletin},
volume = {},
number = {},
pages = {},
pmid = {39752045},
issn = {1995-8218},
}

@article {pmid39748261,
year = {2025},
author = {Wu, JY and Zhang, JY and Xia, WQ and Kang, YN and Liao, RY and Chen, YL and Li, XM and Wen, Y and Meng, FX and Xu, LL and Wen, SH and Liu, HF and Li, YQ and Gu, JR and Lv, Q and Ren, Y},
title = {Predicting autoimmune thyroiditis in primary Sjogren's syndrome patients using a random forest classifier: a retrospective study.},
journal = {Arthritis research & therapy},
volume = {27},
number = {1},
pages = {1},
pmid = {39748261},
issn = {1478-6362},
support = {No. JCYJ20220530145001002//Shenzhen Science and Technology Program/ ; No. JCYJ20220530145001002//Shenzhen Science and Technology Program/ ; No. JCYJ20220530145001002//Shenzhen Science and Technology Program/ ; No. JCYJ20220530145001002//Shenzhen Science and Technology Program/ ; No. JCYJ20220530145001002//Shenzhen Science and Technology Program/ ; No. JCYJ20220530145001002//Shenzhen Science and Technology Program/ ; No. JCYJ20220530145001002//Shenzhen Science and Technology Program/ ; No. JCYJ20220530145001002//Shenzhen Science and Technology Program/ ; No. JCYJ20220530145001002//Shenzhen Science and Technology Program/ ; No. JCYJ20220530145001002//Shenzhen Science and Technology Program/ ; No. JCYJ20220530145001002//Shenzhen Science and Technology Program/ ; No. JCYJ20220530145001002//Shenzhen Science and Technology Program/ ; No. JCYJ20220530145001002//Shenzhen Science and Technology Program/ ; No. JCYJ20220530145001002//Shenzhen Science and Technology Program/ ; No. JCYJ20220530145001002//Shenzhen Science and Technology Program/ ; No. JCYJ20220530145001002//Shenzhen Science and Technology Program/ ; },
mesh = {Humans ; Retrospective Studies ; *Sjogren's Syndrome/immunology/diagnosis ; Female ; Male ; Middle Aged ; *Thyroiditis, Autoimmune/immunology/diagnosis ; Adult ; *Autoantibodies/immunology/blood ; *Machine Learning ; Algorithms ; Aged ; Random Forest ; },
abstract = {BACKGROUND: Primary Sjogren's syndrome (pSS) and autoimmune thyroiditis (AIT) share overlapping genetic and immunological profiles. This retrospective study evaluates the efficacy of machine learning algorithms, with a focus on the Random Forest Classifier, to predict the presence of thyroid-specific autoantibodies (TPOAb and TgAb) in pSS patients.

METHODS: A total of 96 patients with pSS were included in the retrospective study. All participants underwent a complete clinical and laboratory evaluation. All participants underwent thyroid function tests, including TPOAb and TgAb, and were accordingly divided into positive and negative thyroid autoantibody groups. Four machine learning algorithms were then used to analyze the risk factors affecting patients with pSS with positive and negative for thyroid autoantibodies.

RESULTS: The results indicated that the Random Forest Classifier algorithm (AUC = 0.755) outperformed the other three machine learning algorithms. The random forest classifier indicated Age, IgG, C4 and dry mouth were the main factors influencing the prediction of positive thyroid autoantibodies in pSS patients. It is feasible to predict AIT in pSS using machine learning algorithms.

CONCLUSIONS: Analyzing clinical and laboratory data from 96 pSS patients, the Random Forest model demonstrated superior performance (AUC = 0.755), identifying age, IgG levels, complement component 4 (C4), and absence of dry mouth as primary predictors. This approach offers a promising tool for early identification and management of AIT in pSS patients.

TRIAL REGISTRATION: This retrospective study was approved and monitored by the Ethics Committee of The Third Affiliated Hospital of Sun Yat-sen University (No.II2023-254-02).},
}

Humans
Retrospective Studies
*Sjogren's Syndrome/immunology/diagnosis
Female
Male
Middle Aged
*Thyroiditis, Autoimmune/immunology/diagnosis
Adult
*Autoantibodies/immunology/blood
*Machine Learning
Algorithms
Aged
Random Forest

@article {pmid39748063,
year = {2025},
author = {Wei, Q and Li, C and Wang, Y and Gao, X},
title = {Enhancing the performance of SSVEP-based BCIs by combining task-related component analysis and deep neural network.},
journal = {Scientific reports},
volume = {15},
number = {1},
pages = {365},
pmid = {39748063},
issn = {2045-2322},
support = {62066028//National Natural Science Foundation of China/ ; },
mesh = {*Brain-Computer Interfaces ; Humans ; *Evoked Potentials, Visual/physiology ; *Neural Networks, Computer ; *Algorithms ; *Electroencephalography/methods ; Deep Learning ; Signal Processing, Computer-Assisted ; },
abstract = {Steady-State Visually Evoked Potential (SSVEP) signals can be decoded by either a traditional machine learning algorithm or a deep learning network. Combining the two methods is expected to enhance the performance of an SSVEP-based brain-computer interface (BCI) by exploiting their advantages. However, an efficient strategy for integrating the two methods has not yet been established. To address this issue, we propose a classification framework named eTRCA + sbCNN that combines an ensemble task-related component analysis (eTRCA) algorithm and a sub-band convolutional neural network (sbCNN) for recognizing the frequency of SSVEP signals. The two models are first trained separately, then their classification score vectors are added together, and finally the frequency corresponding to the maximal summed score is decided as the frequency of SSVEP signals. The proposed framework can effectively exploit the complementarity between the two kinds of feature signals and significantly improve the classification performance of SSVEP-based BCIs. The performance of the proposed method is validated on two SSVEP BCI datasets and compared with that of eTRCA, sbCNN and other state-of-the-art models. Experimental results indicate that the proposed method significantly outperform the compared algorithms, and thus helps to promote the practical application of SSVEP- BCI systems.},
}

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

@article {pmid39747930,
year = {2025},
author = {Anderson, L and De Ridder, D and Glue, P and Mani, R and van Sleeuwen, C and Smith, M and Adhia, DB},
title = {A safety and feasibility randomized placebo controlled trial exploring electroencephalographic effective connectivity neurofeedback treatment for fibromyalgia.},
journal = {Scientific reports},
volume = {15},
number = {1},
pages = {209},
pmid = {39747930},
issn = {2045-2322},
mesh = {Humans ; *Fibromyalgia/therapy/physiopathology ; *Neurofeedback/methods ; Female ; Middle Aged ; *Electroencephalography ; Adult ; *Feasibility Studies ; Male ; Treatment Outcome ; Gyrus Cinguli/physiopathology/diagnostic imaging ; Chronic Pain/therapy/physiopathology ; },
abstract = {Fibromyalgia is a chronic pain condition contributing to significant disability worldwide. Neuroimaging studies identify abnormal effective connectivity between cortical areas responsible for descending pain modulation (pregenual anterior cingulate cortex, pgACC) and sensory components of pain experience (primary somatosensory cortex, S1). Neurofeedback, a brain-computer interface technique, can normalise dysfunctional brain activity, thereby improving pain and function. This study evaluates the safety, feasibility, and acceptability of a novel electroencephalography-based neurofeedback training, targeting effective alpha-band connectivity from the pgACC to S1 and exploring its effect on pain and function. Participants with fibromyalgia (N = 30; 15 = active, 15 = placebo) received 12 sessions of neurofeedback. Feasibility and outcome measures of pain (Brief Pain Inventory) and function (Revised Fibromyalgia Impact Questionnaire) were collected at baseline and immediately, ten-days, and one-month post-intervention. Descriptive statistics demonstrate effective connectivity neurofeedback training is feasible (recruitment rate: 6 participants per-month, mean adherence: 80.5%, dropout rate: 20%), safe (no adverse events) and highly acceptable (average 8.0/10) treatment approach for fibromyalgia. Active and placebo groups were comparable in their decrease in pain and functional impact. Future fully powered clinical trial is warranted to test the efficacy of the effective connectivity neurofeedback training in people with fibromyalgia with versus without chronic fatigue.},
}

Humans
*Fibromyalgia/therapy/physiopathology
*Neurofeedback/methods
Female
Middle Aged
*Electroencephalography
Adult
*Feasibility Studies
Male
Treatment Outcome
Gyrus Cinguli/physiopathology/diagnostic imaging
Chronic Pain/therapy/physiopathology

@article {pmid39745941,
year = {2025},
author = {Stavisky, SD},
title = {Restoring Speech Using Brain-Computer Interfaces.},
journal = {Annual review of biomedical engineering},
volume = {},
number = {},
pages = {},
doi = {10.1146/annurev-bioeng-110122-012818},
pmid = {39745941},
issn = {1545-4274},
abstract = {People who have lost the ability to speak due to neurological injuries would greatly benefit from assistive technology that provides a fast, intuitive, and naturalistic means of communication. This need can be met with brain-computer interfaces (BCIs): medical devices that bypass injured parts of the nervous system and directly transform neural activity into outputs such as text or sound. BCIs for restoring movement and typing have progressed rapidly in recent clinical trials; speech BCIs are the next frontier. This review covers the clinical need for speech BCIs, surveys foundational studies that point to where and how speech can be decoded in the brain, describes recent progress in both discrete and continuous speech decoding and closed-loop speech BCIs, provides metrics for assessing these systems' performance, and highlights key remaining challenges on the road toward clinically useful speech neuroprostheses.},
}

@article {pmid39745924,
year = {2025},
author = {Trevelyan, AJ and Marks, VS and Graham, RT and Denison, T and Jackson, A and Smith, EH},
title = {On brain stimulation in epilepsy.},
journal = {Brain : a journal of neurology},
volume = {},
number = {},
pages = {},
doi = {10.1093/brain/awae385},
pmid = {39745924},
issn = {1460-2156},
abstract = {Brain stimulation has, for many decades, been considered as a potential solution for the unmet needs of the many people living with drug-resistant epilepsy. Clinically, there are several different approaches in use, including vagus nerve stimulation (VNS), deep brain stimulation of the thalamus, and responsive neurostimulation (RNS). Across populations of patients, all deliver reductions in seizure load and SUDEP risk, yet do so variably, and the improvements seem incremental rather than transformative. In contrast, within the field of experimental neuroscience, the transformational impact of optogenetic stimulation is evident; by providing a means to control subsets of neurons in isolation, it has revolutionized our ability to dissect out the functional relations within neuronal microcircuits. It is worth asking, therefore, how pre-clinical optogenetics research could advance clinical practice in epilepsy? Here, we review the state of the clinical field, and the recent progress in pre-clinical animal research. We report various breakthrough results, including the development of new models of seizure initiation, its use for seizure prediction, and for fast, closed-loop control of pathological brain rhythms, and what these experiments tell us about epileptic pathophysiology. Finally, we consider how these pre-clinical research advances may be translated into clinical practice.},
}

@article {pmid39745545,
year = {2025},
author = {Romano, V and Manto, M},
title = {How and where Effectively Apply Cerebellum Stimulation: The frequency-dependent Modulation of Cerebellar Output by Transcranial Alternating Current Stimulation.},
journal = {Cerebellum (London, England)},
volume = {24},
number = {1},
pages = {22},
pmid = {39745545},
issn = {1473-4230},
mesh = {*Transcranial Direct Current Stimulation/methods ; *Cerebellum/physiology ; Animals ; Humans ; Rats ; },
abstract = {As brain-machine interfaces (BMI) are growingly used in clinical settings, understanding how to apply brain stimulation is increasingly important. Despite the emergence of optogenetic techniques, ethical and medical concerns suggest that interventions that are safe and non-invasive, such as Transcranial Alternating Current Stimulation (tACS), are more likely to be employed in human in the near future. Consequently, the question of how and where to apply current stimulation is becoming increasingly important for the efficient neuromodulation of both neurological and psychiatric disorders. In this edition of The Cerebellum, Mourra et al. demonstrate how ctACS influences cerebellar output at both single-cell and population levels by stimulating Crus I in rats. As the neuron generating this output serves as a crucial convergence and divergence center in the nervous system, it can be leveraged as a strategic hub to target multiple brain structures and influence various behaviors. Accordingly, the discovery that neurons in this relatively deep brain region can be indirectly entrained through Purkinje neuron activation and optimal frequency around 80 Hz could be highly relevant for future medical interventions. In light of these findings, high-γ-tACS might be more effective in humans compared to the more commonly used low-γ (50 Hz) or θ-tACS (5 Hz). This could enhance the chance of cerebellar tACS being utilized in clinical settings and BMI.},
}

*Transcranial Direct Current Stimulation/methods
*Cerebellum/physiology
Animals
Humans
Rats

@article {pmid39744695,
year = {2025},
author = {Li, C and Miao, C and Ge, Y and Wu, J and Gao, P and Yin, S and Zhang, P and Yang, H and Tian, B and Chen, W and Chen, XQ},
title = {A molecularly distinct cell type in the midbrain regulates intermale aggression behaviors in mice.},
journal = {Theranostics},
volume = {15},
number = {2},
pages = {707-725},
pmid = {39744695},
issn = {1838-7640},
mesh = {Animals ; *Aggression/physiology ; Mice ; *Periaqueductal Gray/metabolism/physiology ; Male ; *Neurons/metabolism/physiology ; Mice, Inbred C57BL ; Tachykinins/metabolism/genetics ; Behavior, Animal/physiology ; Mesencephalon/metabolism/physiology ; Serotonin/metabolism ; },
abstract = {Rationale: The periaqueductal gray (PAG) is a central hub for the regulation of aggression, whereas the circuitry and molecular mechanisms underlying this regulation remain uncharacterized. In this study, we investigate the role of a distinct cell type, Tachykinin 2-expressing (Tac2[+]) neurons, located in the dorsomedial PAG (dmPAG) and their modulation of aggressive behavior in mice. Methods: We combined activity mapping, in vivo Ca[2+] recording, chemogenetic and pharmacological manipulation, and a viral-based translating ribosome affinity purification (TRAP) profiling using a mouse resident-intruder model. Results: We revealed that dmPAG[Tac2] neurons are selectively activated by fighting behaviors. Chemogenetic activation of these neurons evoked fighting behaviors, while inhibition or genetic ablation of dmPAG[Tac2] neurons attenuated fighting behaviors. TRAP profiling of dmPAG[Tac2] neurons revealed an enrichment of serotonin-associated transcripts in response to fighting behaviors. Finally, we validated these effects by selectively administering pharmacological agents to the dmPAG, reversing the behavioral outcomes induced by chemogenetic manipulation. Conclusions: We identify dmPAG[Tac2] neurons as critical modulators of aggressive behavior in mouse and thus suggest a distinct molecular target for the treatment of exacerbated aggressive behaviors in populations that exhibit high-level of violence.},
}

Animals
*Aggression/physiology
Mice
*Periaqueductal Gray/metabolism/physiology
Male
*Neurons/metabolism/physiology
Mice, Inbred C57BL
Tachykinins/metabolism/genetics
Behavior, Animal/physiology
Mesencephalon/metabolism/physiology
Serotonin/metabolism

@article {pmid39743292,
year = {2024},
author = {Huang, T and Guo, X and Huang, X and Yi, C and Cui, Y and Dong, Y},
title = {Input-output specific orchestration of aversive valence in lateral habenula during stress dynamics.},
journal = {Journal of Zhejiang University. Science. B},
volume = {25},
number = {12},
pages = {1055-1065},
pmid = {39743292},
issn = {1862-1783},
support = {2022ZD0211700//the STI2030-Major Projects/ ; 32371057, 31922031, 32071017, 81971309, 32170980, 82201707 and 82200562//the National Natural Science Foundation of China/ ; LDQ24C090001//the Zhejiang Provincial Natural Science Foundation of China/ ; 2023-PT310-01//the Non-profit Central Research Institute Fund of Chinese Academy of Medical Sciences/ ; JCTD-2022-11//the CAS Youth Interdisciplinary Team/ ; BX20230319//the China Postdoctoral Science Foundation/ ; 2022B1515020012 and 2021A1515110121//the Guangdong Basic and Applied Basic Research Foundation/ ; 2023B1212060018//the Science and Technology Planning Project of Guangdong Province/ ; JCYJ20210324123212035, RCYX20200714114644167, ZDSYS20220606100801003, JCYJ20210324122809025 and JCYJ20230807110315031//the Shenzhen Fundamental Research Program/ ; },
mesh = {*Habenula/physiology ; Animals ; *Stress, Psychological/physiopathology ; Humans ; Neurons/physiology ; Neuronal Plasticity/physiology ; Depression/physiopathology ; Neural Pathways/physiology ; },
abstract = {Stress has been considered as a major risk factor for depressive disorders, triggering depression onset via inducing persistent dysfunctions in specialized brain regions and neural circuits. Among various regions across the brain, the lateral habenula (LHb) serves as a critical hub for processing aversive information during the dynamic process of stress accumulation, thus having been implicated in the pathogenesis of depression. LHb neurons integrate aversive valence conveyed by distinct upstream inputs, many of which selectively innervate the medial part (LHbM) or lateral part (LHbL) of LHb. LHb subregions also separately assign aversive valence via dissociable projections to the downstream targets in the midbrain which provides feedback loops. Despite these strides, the spatiotemporal dynamics of LHb-centric neural circuits remain elusive during the progression of depression-like state under stress. In this review, we attempt to describe a framework in which LHb orchestrates aversive valence via the input-output specific neuronal architecture. Notably, a physiological form of Hebbian plasticity in LHb under multiple stressors has been unveiled to incubate neuronal hyperactivity in an input-specific manner, which causally encodes chronic stress experience and drives depression onset. Collectively, the recent progress and future efforts in elucidating LHb circuits shed light on early interventions and circuit-specific antidepressant therapies.},
}

*Habenula/physiology
Animals
*Stress, Psychological/physiopathology
Humans
Neurons/physiology
Neuronal Plasticity/physiology
Depression/physiopathology
Neural Pathways/physiology

@article {pmid39743145,
year = {2024},
author = {Hao, X and Ma, M and Meng, F and Liang, H and Liang, C and Liu, X and Zhang, B and Ju, Y and Liu, S and Ming, D},
title = {Diminished attention network activity and heightened salience-default mode transitions in generalized anxiety disorder: Evidence from resting-state EEG microstate analysis.},
journal = {Journal of affective disorders},
volume = {373},
number = {},
pages = {227-236},
doi = {10.1016/j.jad.2024.12.095},
pmid = {39743145},
issn = {1573-2517},
abstract = {Generalized anxiety disorder (GAD) is a common anxiety disorder characterized by excessive, uncontrollable worry and physical symptoms such as difficulty concentrating and sleep disturbances. Although functional magnetic resonance imaging (fMRI) studies have reported aberrant network-level activity related to cognition and emotion in GAD, its low temporal resolution restricts its ability to capture the rapid neural activity in mental processes. EEG microstate analysis offers millisecond-resolution for tracking the dynamic changes in brain electrical activity, thereby illuminating the neurophysiological mechanisms underlying the cognitive and emotional dysfunctions in GAD. This study collected 64-channel resting-state EEG data from 28 GAD patients and 28 healthy controls (HC), identifying five microstate classes (A-E) in both groups. Results showed that GAD patients exhibited significantly lower duration (p < 0.01), occurrence (p < 0.05), and coverage (p < 0.01) of microstate class D, potentially reflecting deficits in attention-related networks. Such alterations may contribute to the impairments in attention maintenance and cognitive control. Additionally, GAD patients displayed reduced transition probabilities in A → D, B → D, C → D, and E → D (all corrected p < 0.05), but increased in C → E (corrected p < 0.05) and E → C (corrected p < 0.01). These results highlight a significant reduction in the brain's ability to transition into microstate class D, alongside overactivity in switching between the default mode network and the salience network. Such neurophysiological changes may underlie cognitive control deficits, increased spontaneous rumination, and emotional regulation challenges observed in GAD. Together, these insights provide a new perspective for understanding the neurophysiological and pathological mechanisms underlying GAD.},
}

@article {pmid39742538,
year = {2024},
author = {Cao, L and Zhao, W and Sun, B},
title = {Emotion recognition using multi-scale EEG features through graph convolutional attention network.},
journal = {Neural networks : the official journal of the International Neural Network Society},
volume = {184},
number = {},
pages = {107060},
doi = {10.1016/j.neunet.2024.107060},
pmid = {39742538},
issn = {1879-2782},
abstract = {Emotion recognition via electroencephalogram (EEG) signals holds significant promise across various domains, including the detection of emotions in patients with consciousness disorders, assisting in the diagnosis of depression, and assessing cognitive load. This process is critically important in the development and research of brain-computer interfaces, where precise and efficient recognition of emotions is paramount. In this work, we introduce a novel approach for emotion recognition employing multi-scale EEG features, denominated as the Dynamic Spatial-Spectral-Temporal Network (DSSTNet). DSSTNet includes three main parts, the first is spatial features extractor, which converts EEG signal into graph structure data, and uses graph convolutional network (GCN) to dynamically optimize the adjacency matrix during the training process to obtain the spatial features between the channels. Next, band attention module is composed of semi-global pooling, localized cross-band interaction and adaptive weighting, which further extracts frequency information. Finally, through the temporal features extractor, the deep temporal information is extracted by stacking several one-dimensional convolutional layers. In addition, in order to improve the performance of emotion recognition and filter valid channels, we add a ℓ2,1-norm regularization term to the loss function to make the adjacency matrix constraint sparse. This makes it easier to preserve emotionally relevant channels and eliminate noise in irrelevant channel. Combined with the channel selection method of graph theory, a small number of optimal channels are selected. We used a self-constructed dataset TJU-EmoEEG and a publicly available SEED dataset to evaluate DSSTNet. The experimental results demonstrate that DSSTNet outperforms current state-of-the-art (SOTA) methods in emotional recognition tasks.},
}

@article {pmid39741784,
year = {2024},
author = {Degirmenci, M and Yuce, YK and Perc, M and Isler, Y},
title = {EEG channel and feature investigation in binary and multiple motor imagery task predictions.},
journal = {Frontiers in human neuroscience},
volume = {18},
number = {},
pages = {1525139},
pmid = {39741784},
issn = {1662-5161},
abstract = {INTRODUCTION: Motor Imagery (MI) Electroencephalography (EEG) signals are non-stationary and dynamic physiological signals which have low signal-to-noise ratio. Hence, it is difficult to achieve high classification accuracy. Although various machine learning methods have already proven useful to that effect, the use of many features and ineffective EEG channels often leads to a complex structure of classifier algorithms. State-of-the-art studies were interested in improving classification performance with complex feature extraction and classification methods by neglecting detailed EEG channel and feature investigation in predicting MI tasks from EEGs. Here, we investigate the effects of the statistically significant feature selection method on four different feature domains (time-domain, frequency-domain, time-frequency domain, and non-linear domain) and their two different combinations to reduce the number of features and classify MI-EEG features by comparing low-dimensional matrices with well-known machine learning algorithms.

METHODS: Our main goal is not to find the best classifier performance but to perform feature and channel investigation in MI task classification. Therefore, the detailed investigation of the effect of EEG channels and features is implemented using a statistically significant feature distribution on 22 EEG channels for each feature set separately. We used the BCI Competition IV Dataset IIa and 288 samples per person. A total of 1,364 MI-EEG features were analyzed in this study. We tested nine distinct classifiers: Decision tree, Discriminant analysis, Logistic regression, Naive Bayes, Support vector machine, k-Nearest neighbor, Ensemble learning, Neural networks, and Kernel approximation.

RESULTS: Among all feature sets considered, classifications performed with non-linear and combined feature sets resulted in a maximum accuracy of 63.04% and 47.36% for binary and multiple MI task predictions, respectively. The ensemble learning classifier achieved the maximum accuracy in almost all feature sets for binary and multiple MI task classifications.

DISCUSSION: Our research thus shows that the statistically significant feature-based feature selection method significantly improves the classification performance with fewer features in almost all feature sets, enabling detailed and effective EEG channel and feature investigation.},
}

@article {pmid39741250,
year = {2024},
author = {Liang, S and Li, L and Zu, W and Feng, W and Hang, W},
title = {Adaptive deep feature representation learning for cross-subject EEG decoding.},
journal = {BMC bioinformatics},
volume = {25},
number = {1},
pages = {393},
pmid = {39741250},
issn = {1471-2105},
support = {61902197//National Natural Science Foundation of China/ ; KYCX23_1073//Postgraduate Research & Practice Innovation Program of Jiangsu Province/ ; 23KJB520012//Natural Science Research of Jiangsu Higher Education Institutions of China/ ; },
mesh = {*Electroencephalography/methods ; Humans ; Deep Learning ; Algorithms ; Signal Processing, Computer-Assisted ; Brain-Computer Interfaces ; },
abstract = {BACKGROUND: The collection of substantial amounts of electroencephalogram (EEG) data is typically time-consuming and labor-intensive, which adversely impacts the development of decoding models with strong generalizability, particularly when the available data is limited. Utilizing sufficient EEG data from other subjects to aid in modeling the target subject presents a potential solution, commonly referred to as domain adaptation. Most current domain adaptation techniques for EEG decoding primarily focus on learning shared feature representations through domain alignment strategies. Since the domain shift cannot be completely removed, target EEG samples located near the edge of clusters are also susceptible to misclassification.

METHODS: We propose a novel adaptive deep feature representation (ADFR) framework to improve the cross-subject EEG classification performance through learning transferable EEG feature representations. Specifically, we first minimize the distribution discrepancy between the source and target domains by employing maximum mean discrepancy (MMD) regularization, which aids in learning the shared feature representations. We then utilize the instance-based discriminative feature learning (IDFL) regularization to make the learned feature representations more discriminative. Finally, the entropy minimization (EM) regularization is further integrated to adjust the classifier to pass through the low-density region between clusters. The synergistic learning between above regularizations during the training process enhances EEG decoding performance across subjects.

RESULTS: The effectiveness of the ADFR framework was evaluated on two public motor imagery (MI)-based EEG datasets: BCI Competition III dataset 4a and BCI Competition IV dataset 2a. In terms of average accuracy, ADFR achieved improvements of 3.0% and 2.1%, respectively, over the state-of-the-art methods on these datasets.

CONCLUSIONS: The promising results highlight the effectiveness of the ADFR algorithm for EEG decoding and show its potential for practical applications.},
}

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

@article {pmid39738346,
year = {2024},
author = {Le, TT and Luong, DAQ and Joo, H and Kim, D and Woo, J},
title = {Differences in spatiotemporal dynamics for processing specific semantic categories: An EEG study.},
journal = {Scientific reports},
volume = {14},
number = {1},
pages = {31918},
pmid = {39738346},
issn = {2045-2322},
support = {NRF-2020R1A2C2003319//National Research Foundation of Korea/ ; },
mesh = {Humans ; *Electroencephalography ; *Semantics ; Male ; Female ; Adult ; *Brain Mapping ; Brain/physiology ; Young Adult ; Comprehension/physiology ; },
abstract = {Semantic processing is an essential mechanism in human language comprehension and has profound implications for speech brain-computer interface technologies. Despite recent advances in brain imaging techniques and data analysis algorithms, the mechanisms underlying human brain semantic representations remain a topic of debate, specifically whether this occurs through the activation of selectively separated cortical regions or via a network of distributed and overlapping regions. This study investigates spatiotemporal neural representation during the perception of semantic words related to faces, numbers, and animals using electroencephalography. Source-level analysis focuses on contrasting neural responses to different semantic categories. Critical intervals used in the source contrast analysis are defined using the peak duration of global field power. Effective connectivity, determined through a causality analysis of brain regions activated for semantic processing, is explored. The findings reveal the necessity of a distributed network of regions for processing specific semantic categories and provide evidence suggesting the existence of a neural substrate for semantic representations.},
}

Humans
*Electroencephalography
*Semantics
Male
Female
Adult
*Brain Mapping
Brain/physiology
Young Adult
Comprehension/physiology

@article {pmid39738221,
year = {2024},
author = {Wolde, HF and Clements, ACA and Gilmour, B and Alene, KA},
title = {Spatial co-distribution of tuberculosis prevalence and low BCG vaccination coverage in Ethiopia.},
journal = {Scientific reports},
volume = {14},
number = {1},
pages = {31561},
pmid = {39738221},
issn = {2045-2322},
support = {APP1196549//Australian National Health and Medical Research Council/ ; },
mesh = {Humans ; Ethiopia/epidemiology ; *BCG Vaccine/administration & dosage ; Prevalence ; *Vaccination Coverage/statistics & numerical data ; Female ; *Tuberculosis/epidemiology/prevention & control ; Male ; *Bayes Theorem ; Infant ; Child, Preschool ; Adult ; Adolescent ; Vaccination/statistics & numerical data ; Young Adult ; Child ; Spatial Analysis ; },
abstract = {While bacille-calmette-guerin (BCG) vaccination is one of the recommended strategies for preventing tuberculosis (TB), its coverage is low in several countries, including Ethiopia. This study investigated the spatial co-distribution and drivers of TB prevalence and low BCG coverage in Ethiopia. This ecological study was conducted using data from a national TB prevalence survey and the Ethiopian demographic and health survey (EDHS) to map the spatial co-distribution of BCG vaccination coverage and TB prevalence. A Bayesian geostatistical model was built to identify the drivers for the spatial distribution of TB prevalence and low BCG vaccination coverage. BCG vaccination coverage was defined as the number of children who received the vaccine divided by the total number of children born within five years preceding the EDHS surveys. Parameter estimation was done using binary logistic regression. Prediction maps for the co-distribution of high TB prevalence and low BCG vaccination coverage were created by overlying spatial prediction surfaces of the two outcomes. Posterior means and a 95% Bayesian credible interval (CrI) were used to summarize the parameters of the model. The national prevalence was 0.40% (95% confidence interval (CI) 0.34%, 0.47%) for TB and 47% (95% CI 46%, 48%) for vaccination coverage. Substantial spatial variation in TB prevalence and low BCG coverage was observed at a regional and local level, particularly in border areas of the country, including the Somali, Afar, and Oromia regions. Approximately 58% of the pixels (i.e., geographical area or spatial units) with high TB prevalence exhibited low BCG coverage in the same location. While travel time to cities (Mean = 0.28, 95% BCI: 0.15, 0.41) and distance to health facilities (Mean = 0.43, 95% CI 0.22, 0.63), were positively associated, population density (Mean = -0.04, 95% BCI -0.05, -0.02) was negatively associated, with the proportion of unvaccinated children for BCG indicating areas near health facilities and cities have better BCG coverage. However, there were no significant predictors for TB prevalence. Substantial spatial co-distribution between high TB prevalence and low BCG coverage was observed in some parts of the country, indicating that there are areas where the TB burden is not being adequately managed through the provision of vaccines in Ethiopia. Scaling up BCG vaccination coverage and TB diagnosis and treatment through improving access to health services in border regions such as Somalia and Afar would be important to reduce the prevalence of TB in Ethiopia.},
}

Humans
Ethiopia/epidemiology
*BCG Vaccine/administration & dosage
Prevalence
*Vaccination Coverage/statistics & numerical data
Female
*Tuberculosis/epidemiology/prevention & control
Male
*Bayes Theorem
Infant
Child, Preschool
Adult
Adolescent
Vaccination/statistics & numerical data
Young Adult
Child
Spatial Analysis

@article {pmid39738213,
year = {2024},
author = {Liu, Y and Huang, S and Xu, W and Wang, Z and Ming, D},
title = {An fMRI study on the generalization of motor learning after brain actuated supernumerary robot training.},
journal = {NPJ science of learning},
volume = {9},
number = {1},
pages = {80},
pmid = {39738213},
issn = {2056-7936},
support = {62273251//National Natural Science Foundation of China (National Science Foundation of China)/ ; 81925020//National Natural Science Foundation of China (National Science Foundation of China)/ ; 21JCYBJC00520//Natural Science Foundation of Tianjin City (Natural Science Foundation of Tianjin)/ ; MSV202418//State Key Laboratory of Mechanical System and Vibration/ ; },
abstract = {Generalization is central to motor learning. However, few studies are on the learning generalization of BCI-actuated supernumerary robotic finger (BCI-SRF) for human-machine interaction training, and no studies have explored its longitudinal neuroplasticity mechanisms. Here, 20 healthy right-handed participants were recruited and randomly assigned to BCI-SRF group or inborn finger group (Finger) for 4-week training and measured by novel SRF-finger opposition sequences and multimodal MRI. After training, the BCI-SRF group showed 350% times compared to the Finger group in the improvement of sequence opposition accuracy before and after training, and accompanied by significant functional connectivity increases in the sensorimotor region and prefrontal cortex, as well as in the intra- and inter-hemisphere of the sensorimotor network. Moreover, Granger Causality Analysis identified causal effect main transfer within the sensorimotor cortex-cerebellar-thalamus loop and frontal-parietal loop. The findings suggest that BCI-SRF training enhances motor sequence learning ability by influencing the functional reorganization of sensorimotor network.},
}

@article {pmid39735964,
year = {2024},
author = {Luo, TJ and Li, J and Li, R and Zhang, X and Wu, SR and Peng, H},
title = {Motion Cognitive Decoding of Cross-Subject Motor Imagery Guided on Different Visual Stimulus Materials.},
journal = {Journal of integrative neuroscience},
volume = {23},
number = {12},
pages = {218},
doi = {10.31083/j.jin2312218},
pmid = {39735964},
issn = {0219-6352},
support = {24JCXK01YB//Planning Project of Philosophy and Social Science of Zhejiang Province/ ; 24JCXK02YB//Planning Project of Philosophy and Social Science of Zhejiang Province/ ; 62106049//National Natural Science Foundation of China/ ; 61662025//National Natural Science Foundation of China/ ; 61871289//National Natural Science Foundation of China/ ; 62007016//National Natural Science Foundation of China/ ; 2022J01655//Natural Science Foundation of Fujian Province of China/ ; },
mesh = {Humans ; *Imagination/physiology ; *Electroencephalography/methods ; Adult ; Young Adult ; Male ; *Brain-Computer Interfaces ; Female ; Motor Activity/physiology ; Psychomotor Performance/physiology ; Motion Perception/physiology ; },
abstract = {BACKGROUND: Motor imagery (MI) plays an important role in brain-computer interfaces, especially in evoking event-related desynchronization and synchronization (ERD/S) rhythms in electroencephalogram (EEG) signals. However, the procedure for performing a MI task for a single subject is subjective, making it difficult to determine the actual situation of an individual's MI task and resulting in significant individual EEG response variations during motion cognitive decoding.

METHODS: To explore this issue, we designed three visual stimuli (arrow, human, and robot), each of which was used to present three MI tasks (left arm, right arm, and feet), and evaluated differences in brain response in terms of ERD/S rhythms. To compare subject-specific variations of different visual stimuli, a novel cross-subject MI-EEG classification method was proposed for the three visual stimuli. The proposed method employed a covariance matrix centroid alignment for preprocessing of EEG samples, followed by a model agnostic meta-learning method for cross-subject MI-EEG classification.

RESULTS AND CONCLUSION: The experimental results showed that robot stimulus materials were better than arrow or human stimulus materials, with an optimal cross-subject motion cognitive decoding accuracy of 79.04%. Moreover, the proposed method produced robust classification of cross-subject MI-EEG signal decoding, showing superior results to conventional methods on collected EEG signals.},
}

Humans
*Imagination/physiology
*Electroencephalography/methods
Adult
Young Adult
Male
*Brain-Computer Interfaces
Female
Motor Activity/physiology
Psychomotor Performance/physiology
Motion Perception/physiology

@article {pmid39733553,
year = {2024},
author = {Jain, A and Kumar, L},
title = {ESI-GAL: EEG source imaging-based trajectory estimation for grasp and lift task.},
journal = {Computers in biology and medicine},
volume = {186},
number = {},
pages = {109608},
doi = {10.1016/j.compbiomed.2024.109608},
pmid = {39733553},
issn = {1879-0534},
abstract = {BACKGROUND: Electroencephalogram (EEG) signals-based motor kinematics prediction (MKP) has been an active area of research to develop Brain-computer interface (BCI) systems such as exosuits, prostheses, and rehabilitation devices. However, EEG source imaging (ESI) based kinematics prediction is sparsely explored in the literature.

METHOD: In this study, pre-movement EEG features are utilized to predict three-dimensional (3D) hand kinematics for the grasp-and-lift motor task. A public dataset, WAY-EEG-GAL, is utilized for MKP analysis. In particular, sensor-domain (EEG data) and source-domain (ESI data) based features from the frontoparietal region are explored for MKP. Deep learning-based models are explored to achieve efficient kinematics decoding. Various time-lagged and window sizes are analyzed for hand kinematics prediction. Subsequently, intra-subject and inter-subject MKP analysis is performed to investigate the subject-specific and subject-independent motor-learning capabilities of the neural decoders. The Pearson correlation coefficient (PCC) is used as the performance metric for kinematics trajectory decoding.

RESULTS: The rEEGNet neural decoder achieved the best performance with sensor-domain and source-domain features with the time lag and window size of 100ms and 450ms, respectively. The highest mean PCC values of 0.790, 0.795, and 0.637 are achieved using sensor-domain features, while 0.769, 0.777, and 0.647 are achieved using source-domain features in x, y, and z-directions, respectively.

CONCLUSION: This study explores the feasibility of trajectory prediction using EEG sensor-domain and source-domain features for the grasp-and-lift task. Furthermore, inter-subject trajectory estimation is performed using the proposed deep learning decoder with EEG source domain features.},
}

@article {pmid39733023,
year = {2024},
author = {Cheng, Z and Bu, X and Wang, Q and Yang, T and Tu, J},
title = {EEG-based emotion recognition using multi-scale dynamic CNN and gated transformer.},
journal = {Scientific reports},
volume = {14},
number = {1},
pages = {31319},
pmid = {39733023},
issn = {2045-2322},
support = {2020CB-34//the Science and Technology Plan Project of Jingzhou City/ ; 2022BCE009//Key Plan of Science and Technology Department of Hubei Province/ ; },
mesh = {*Electroencephalography/methods ; *Emotions/physiology ; Humans ; *Neural Networks, Computer ; Signal Processing, Computer-Assisted ; Brain-Computer Interfaces ; Algorithms ; },
abstract = {Emotions play a crucial role in human thoughts, cognitive processes, and decision-making. EEG has become a widely utilized tool in emotion recognition due to its high temporal resolution, real-time monitoring capabilities, portability, and cost-effectiveness. In this paper, we propose a novel end-to-end emotion recognition method from EEG signals, called MSDCGTNet, which is based on the Multi-Scale Dynamic 1D CNN and the Gated Transformer. First, the Multi-Scale Dynamic CNN is used to extract complex spatial and spectral features from raw EEG signals, which not only avoids information loss but also reduces computational costs associated with the time-frequency conversion of signals. Then, the Gated Transformer Encoder is utilized to capture global dependencies of EEG signals. This encoder focuses on specific regions of the input sequence while reducing computational resources through parallel processing with the improved multi-head self-attention mechanisms. Third, the Temporal Convolution Network is used to extract temporal features from the EEG signals. Finally, the extracted abstract features are fed into a classification module for emotion recognition. The proposed method was evaluated on three publicly available datasets: DEAP, SEED, and SEED_IV. Experimental results demonstrate the high accuracy and efficiency of the proposed method for emotion recognition. This approach proves to be robust and suitable for various practical applications. By addressing challenges posed by existing methods, the proposed method provides a valuable and effective solution for the field of Brain-Computer Interface (BCI).},
}

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

@article {pmid39732819,
year = {2024},
author = {Jin, L and Hu, J and Li, Y and Zhu, Y and He, X and Bai, R and Wang, L},
title = {Altered neurovascular coupling and structure-function coupling in Moyamoya disease affect postoperative collateral formation.},
journal = {Scientific reports},
volume = {14},
number = {1},
pages = {31324},
pmid = {39732819},
issn = {2045-2322},
support = {81870910//National Natural Science Foundation of China/ ; 2022C03133//Key Research and Development Program of Zhejiang Province/ ; },
mesh = {Humans ; *Moyamoya Disease/physiopathology/surgery/diagnostic imaging ; Female ; Male ; Adult ; *Cerebrovascular Circulation/physiology ; *Neurovascular Coupling/physiology ; *Magnetic Resonance Imaging ; Diffusion Tensor Imaging ; Middle Aged ; Collateral Circulation/physiology ; Case-Control Studies ; Brain/physiopathology/diagnostic imaging/pathology ; Young Adult ; },
abstract = {Chronic ischemia in moyamoya disease (MMD) impaired white matter microstructure and neural functional network. However, the coupling between cerebral blood flow (CBF) and functional connectivity and the association between structural and functional network are largely unknown. 38 MMD patients and 20 sex/age-matched healthy controls (HC) were included for T1-weighted imaging, arterial spin labeling imaging, resting-state functional MRI and diffusion tensor imaging. All patients had preoperative and postoperative digital subtraction angiography. Upon constructing the structural connectivity (SC) and functional connectivity (FC) networks, the SC-FC coupling was calculated. After obtaining the graph theoretical parameters, neurovascular coupling represented the spatial correlation between node degree centrality (DC) of functional networks and CBF. The CBF-DC coupling and SC-FC coupling were compared between MMD and HC groups. We further analyzed the correlation between coupling indexes and cognitive scores, as well as postoperative collateral formation. Compared with HC, CBF-DC coupling was decreased in MMD (p = 0.021), especially in the parietal lobe (p = 0.047). SC-FC coupling in MMD decreased in frontal, occipital, and subcortical regions. Cognitive scores were correlated with the CBF-DC coupling in frontal lobes (r = 0.394, p = 0.029) and SC-FC coupling (r = 0.397, p = 0.027). The CBF-DC coupling of patients with good postoperative collateral formation was higher (p = 0.041). Overall, neurovascular decoupling and structure-functional decoupling at the cortical level may be the underlying neuropathological mechanisms of MMD.},
}

Humans
*Moyamoya Disease/physiopathology/surgery/diagnostic imaging
Female
Male
Adult
*Cerebrovascular Circulation/physiology
*Neurovascular Coupling/physiology
*Magnetic Resonance Imaging
Diffusion Tensor Imaging
Middle Aged
Collateral Circulation/physiology
Case-Control Studies
Brain/physiopathology/diagnostic imaging/pathology
Young Adult

@article {pmid39731856,
year = {2024},
author = {Guo, X and Feng, Y and Ji, X and Jia, N and Maimaiti, A and Lai, J and Wang, Z and Yang, S and Hu, S},
title = {Shared genetic architecture and bidirectional clinical risks within the psycho-metabolic nexus.},
journal = {EBioMedicine},
volume = {111},
number = {},
pages = {105530},
doi = {10.1016/j.ebiom.2024.105530},
pmid = {39731856},
issn = {2352-3964},
abstract = {BACKGROUND: Increasing evidence suggests a complex interplay between psychiatric disorders and metabolic dysregulations. However, most research has been limited to specific disorder pairs, leaving a significant gap in our understanding of the broader psycho-metabolic nexus.

METHODS: This study leveraged large-scale cohort data and genome-wide association study (GWAS) summary statistics, covering 8 common psychiatric disorders and 43 metabolic traits. We introduced a comprehensive analytical strategy to identify shared genetic bases sequentially, from key genetic correlation regions to local pleiotropy and pleiotropic genes. Finally, we developed polygenic risk score (PRS) models to translate these findings into clinical applications.

FINDINGS: We identified significant bidirectional clinical risks between psychiatric disorders and metabolic dysregulations among 310,848 participants from the UK Biobank. Genetic correlation analysis confirmed 104 robust trait pairs, revealing 1088 key genomic regions, including critical hotspots such as chr3: 47588462-50387742. Cross-trait meta-analysis uncovered 388 pleiotropic single nucleotide variants (SNVs) and 126 shared causal variants. Among variants, 45 novel SNVs were associated with psychiatric disorders and 75 novel SNVs were associated with metabolic traits, shedding light on new targets to unravel the mechanism of comorbidity. Notably, RBM6, a gene involved in alternative splicing and cellular stress response regulation, emerged as a key pleiotropic gene. When psychiatric and metabolic genetic information were integrated, PRS models demonstrated enhanced predictive power.

INTERPRETATION: The study highlights the intertwined genetic and clinical relationships between psychiatric disorders and metabolic dysregulations, emphasising the need for integrated approaches in diagnosis and treatment.

FUNDING: The National Key Research and Development Program of China (2023YFC2506200, SHH). The National Natural Science Foundation of China (82273741, SY).},
}

@article {pmid39730087,
year = {2024},
author = {Kong, S and Zhang, J and Wang, L and Li, W and Guo, H and Weng, Q and He, Q and Lou, H and Ding, L and Yang, B},
title = {Mechanisms of low MHC I expression and strategies for targeting MHC I with small molecules in cancer immunotherapy.},
journal = {Cancer letters},
volume = {611},
number = {},
pages = {217432},
doi = {10.1016/j.canlet.2024.217432},
pmid = {39730087},
issn = {1872-7980},
abstract = {Major histocompatibility complex (MHC) class I load antigens and present them on the cell surface, which transduces the tumor-associated antigens to CD8[+] T cells, activating the acquired immune system. However, many tumors downregulate MHC I expression to evade immune surveillance. The low expression of MHC I not only reduce recognition by- and cytotoxicity of CD8[+] T cells, but also seriously weakens the anti-tumor effect of immunotherapy by restoring CD8[+] T cells, such as immune checkpoint inhibitors (ICIs). Accumulated evidence suggested that restoring MHC I expression is an effective strategy for enhancing tumor immunotherapy. This review focuses on mechanisms underlying MHC I downregulation include gene deletion and mutation, transcriptional inhibition, reduced mRNA stability, increased protein degradation, and disruption of endocytic trafficking. We also provide a comprehensive review of small molecules that restore or upregulate MHC I expression, as well as clinical trials involving the combination of ICIs and these small molecule drugs.},
}

@article {pmid39729483,
year = {2024},
author = {Gao, S and Sun, Y and Wu, F and Jiang, J and Peng, T and Zhang, R and Ling, C and Han, Y and Xu, Q and Zou, L and Liao, Y and Liang, C and Zhang, D and Qi, S and Tang, J and Xu, X},
title = {Effects on Multimodal Connectivity Patterns in Female Schizophrenia During 8 Weeks of Antipsychotic Treatment.},
journal = {Schizophrenia bulletin},
volume = {},
number = {},
pages = {},
doi = {10.1093/schbul/sbae176},
pmid = {39729483},
issn = {1745-1701},
support = {82172061//National Natural Science Foundation of China/ ; BE2022677//Key Research and Development Plan in Jiangsu/ ; WSN-166//16th Batch of Six Talent Peak Projects in Jiangsu/ ; YKK23138//Nanjing Health Technology Development Project/ ; 23-25-289//Training and Management of Young Talents in Nanjing Brain Hospital/ ; 2016YFC1306900//National Key R&D Program of China/ ; },
abstract = {BACKGROUND AND HYPOTHESIS: Respective abnormal structural connectivity (SC) and functional connectivity (FC) have been reported in individuals with schizophrenia. However, transmodal associations between SC and FC following antipsychotic treatment, especially in female schizophrenia, remain unclear. We hypothesized that increased SC-FC coupling may be found in female schizophrenia, and could be normalized after antipsychotic treatment.

STUDY DESIGN: Sixty-four female drug-naïve patients with first-diagnosed schizophrenia treated with antipsychotic drugs for 8 weeks, and 55 female healthy controls (HCs) were enrolled. Magnetic resonance imaging (MRI) data were collected from HCs at baseline and from patients at baseline and after treatment. SC and FC were analyzed by network-based statistics, calculating nonzero SC-FC coupling of the whole brain and altered connectivity following treatment. Finally, an Elastic-net logistic regression analysis was employed to establish a predictive model for evaluating the clinical efficacy treatment.

STUDY RESULTS: At baseline, female schizophrenia patients exhibited abnormal SC in cortico-cortical, frontal-limbic, frontal-striatal, limbic-striatal, and limbic-cerebellar connectivity compared to HCs, while FC showed no abnormalities. Following treatment, cortico-cortical, frontal-limbic, frontal-striatal, limbic-striatal, temporal-cerebellar, and limbic-cerebellar connectivity were altered in both SC and FC. Additionally, SC-FC coupling of altered connectivity was higher in patients at baseline than in HC, trending toward normalization after treatment. Furthermore, identified FC or/and SC predicted changes in psychopathological symptoms and cognitive impairment among female schizophrenia following treatment.

CONCLUSIONS: SC-FC coupling may be a potential predictive biomarker of treatment response. Cortico-cortical, frontal-limbic, frontal-striatal, limbic-striatal, temporal-cerebellar, and limbic-cerebellar could represent major targets for antipsychotic drugs in female schizophrenia.},
}

@article {pmid39727765,
year = {2024},
author = {Yousefipour, B and Rajabpour, V and Abdoljabbari, H and Sheykhivand, S and Danishvar, S},
title = {An Ensemble Deep Learning Approach for EEG-Based Emotion Recognition Using Multi-Class CSP.},
journal = {Biomimetics (Basel, Switzerland)},
volume = {9},
number = {12},
pages = {},
pmid = {39727765},
issn = {2313-7673},
abstract = {In recent years, significant advancements have been made in the field of brain-computer interfaces (BCIs), particularly in the area of emotion recognition using EEG signals. The majority of earlier research in this field has missed the spatial-temporal characteristics of EEG signals, which are critical for accurate emotion recognition. In this study, a novel approach is presented for classifying emotions into three categories, positive, negative, and neutral, using a custom-collected dataset. The dataset used in this study was specifically collected for this purpose from 16 participants, comprising EEG recordings corresponding to the three emotional states induced by musical stimuli. A multi-class Common Spatial Pattern (MCCSP) technique was employed for the processing stage of the EEG signals. These processed signals were then fed into an ensemble model comprising three autoencoders with Convolutional Neural Network (CNN) layers. A classification accuracy of 99.44 ± 0.39% for the three emotional classes was achieved by the proposed method. This performance surpasses previous studies, demonstrating the effectiveness of the approach. The high accuracy indicates that the method could be a promising candidate for future BCI applications, providing a reliable means of emotion detection.},
}

@article {pmid39727293,
year = {2024},
author = {Gefen, N and Mazer, B and Krasovsky, T and Weiss, PL},
title = {Novel rehabilitation technologies in pediatric rehabilitation: knowledge towards translation.},
journal = {Disability and rehabilitation. Assistive technology},
volume = {},
number = {},
pages = {1-10},
doi = {10.1080/17483107.2024.2445017},
pmid = {39727293},
issn = {1748-3115},
abstract = {Purpose: Knowledge translation (KT) refers to the process of applying the most promising research outcomes into practice to ensure that new discoveries and innovations improve healthcare accessibility, effectiveness, and accountability. The objective of this perspective paper is to discuss and illustrate via examples how the KT process can be implemented in an era of rapid advancement in rehabilitation technologies that have the potential to significantly impact pediatric healthcare. Methods: Using Graham et al.'s (2006) Knowledge-to-Action cycle, which includes the knowledge creation funnel and the action cycle, we illustrate its application in implementing novel technologies into clinical practice and informing healthcare policy changes. We explore three successful applications of technology research: powered mobility, head support systems, and telerehabilitation. Additionally, we examine less clinically mature technologies such as brain-computer interfaces and robotic assistive devices, which are hindered by cost, robustness, and ease-of-use issues. Conclusions: The paper concludes by discussing how technology acceptance and usage in clinical settings are influenced by various barriers and facilitators at different stakeholder levels, including clients, families, clinicians, management, researchers, developers, and society. Recommendations include focusing on early and ongoing design partnerships, transitioning from research to real-life implementation, and identifying optimal timing for clinical adoption of new technologies.},
}

@article {pmid39726882,
year = {2024},
author = {Afrah, R and Amini, Z and Kafieh, R},
title = {An Unsupervised Feature Extraction Method based on CLSTM-AE for Accurate P300 Classification in Brain-Computer Interface Systems.},
journal = {Journal of biomedical physics & engineering},
volume = {14},
number = {6},
pages = {579-592},
pmid = {39726882},
issn = {2251-7200},
abstract = {BACKGROUND: The P300 signal, an endogenous component of event-related potentials, is extracted from an electroencephalography signal and employed in Brain-computer Interface (BCI) devices.

OBJECTIVE: The current study aimed to address challenges in extracting useful features from P300 components and detecting P300 through a hybrid unsupervised manner based on Convolutional Neural Network (CNN) and Long Short-term Memory (LSTM).

MATERIAL AND METHODS: In this cross-sectional study, CNN as a useful method for the P300 classification task emphasizes spatial characteristics of data. However, CNN and LSTM networks are combined to modify the classification system by extracting both spatial and temporal features. Then, the CNN-LSTM network was trained in an unsupervised learning method based on an autoencoder to improve Signal-to-noise Ratio (SNR) by extracting main components from latent space. To deal with imbalanced data, an Adaptive Synthetic Sampling Approach (ADASYN) is used and augmented without any duplication.

RESULTS: The trained model, tested on the BCI competition III dataset, including two normal subjects, with an accuracy of 95% and 94% for subjects A and B in P300 detection, respectively.

CONCLUSION: CNN-LSTM, was embedded into an autoencoder and introduced to simultaneously extract spatial and temporal features and manage the computational complexity of the method. Further, ADASYN as an augmentation method was proposed to deal with the imbalanced nature of data, which not only maintained feature space as before but also preserved anatomical features of P300. High-quality results highlight the suitable efficiency of the proposed method.},
}

@article {pmid39726001,
year = {2024},
author = {Sun, Q and Merino, EC and Yang, L and Van Hulle, MM},
title = {Unraveling EEG correlates of unimanual finger movements: insights from non-repetitive flexion and extension tasks.},
journal = {Journal of neuroengineering and rehabilitation},
volume = {21},
number = {1},
pages = {228},
pmid = {39726001},
issn = {1743-0003},
support = {202206050022//China Scholarship Council/ ; 101118964//Horizon Europe's Marie Sklodowska-Curie Action/ ; 857375//Horizon 2020 research and innovation programme/ ; C24/18/098//special research fund of the KU Leuven/ ; G0A4118N, G0A4321N, G0C1522N//Belgian Fund for Scientific Research - Flanders/ ; AKUL 043//Hercules Foundation/ ; },
mesh = {Humans ; *Fingers/physiology ; Male ; *Electroencephalography/methods ; Female ; Adult ; *Movement/physiology ; Young Adult ; Brain-Computer Interfaces ; },
abstract = {BACKGROUND: The loss of finger control in individuals with neuromuscular disorders significantly impacts their quality of life. Electroencephalography (EEG)-based brain-computer interfaces that actuate neuroprostheses directly via decoded motor intentions can help restore lost finger mobility. However, the extent to which finger movements exhibit distinct and decodable EEG correlates remains unresolved. This study aims to investigate the EEG correlates of unimanual, non-repetitive finger flexion and extension.

METHODS: Sixteen healthy, right-handed participants completed multiple sessions of right-hand finger movement experiments. These included five individual (Thumb, Index, Middle, Ring, and Pinky) and four coordinated (Pinch, Point, ThumbsUp, and Fist) finger flexions and extensions, along with a rest condition (None). High-density EEG and finger trajectories were simultaneously recorded and analyzed. We examined low-frequency (0.3-3 Hz) time series and movement-related cortical potentials (MRCPs), and event-related desynchronization/synchronization (ERD/S) in the alpha- (8-13 Hz) and beta (13-30 Hz) bands. A clustering approach based on Riemannian distances was used to chart similarities between the broadband EEG responses (0.3-70 Hz) to the different finger scenarios. The contribution of different state-of-the-art features was identified across sub-bands, from low-frequency to low gamma (30-70 Hz), and an ensemble approach was used to pairwise classify single-trial finger movements and rest.

RESULTS: A significant decrease in EEG amplitude in the low-frequency time series was observed in the contralateral frontal-central regions during finger flexion and extension. Distinct MRCP patterns were found in the pre-, ongoing-, and post-movement stages. Additionally, strong ERD was detected in the contralateral central brain regions in both alpha and beta bands during finger flexion and extension, with the beta band showing a stronger rebound (ERS) post-movement. Within the finger movement repertoire, the Thumb was most distinctive, followed by the Fist. Decoding results indicated that low-frequency time-domain amplitude better differentiates finger movements, while alpha and beta band power and Riemannian features better detect movement versus rest. Combining these features yielded over 80% finger movement detection accuracy, while pairwise classification accuracy exceeded 60% for the Thumb versus the other fingers.

CONCLUSION: Our findings confirm that non-repetitive finger movements, whether individual or coordinated, can be precisely detected from EEG. However, differentiating between specific movements is challenging due to highly overlapping neural correlates in time, spectral, and spatial domains. Nonetheless, certain finger movements, such as those involving the Thumb, exhibit distinct EEG responses, making them prime candidates for dexterous finger neuroprostheses.},
}

Humans
*Fingers/physiology
Male
*Electroencephalography/methods
Female
Adult
*Movement/physiology
Young Adult
Brain-Computer Interfaces

@article {pmid39725763,
year = {2024},
author = {Fu, R and Niu, S and Feng, X and Shi, Y and Jia, C and Zhao, J and Wen, G},
title = {Performance investigation of MVMD-MSI algorithm in frequency recognition for SSVEP-based brain-computer interface and its application in robotic arm control.},
journal = {Medical & biological engineering & computing},
volume = {},
number = {},
pages = {},
pmid = {39725763},
issn = {1741-0444},
support = {62073282//the National Natural Science Foundation of China/ ; F2022203092//Natural Science Foundation of Hebei Province/ ; F2020203070//Natural Science Foundation of Hebei Province/ ; 206Z0301G//the Central Guidance on Local Science and Technology Development Fund of Hebei Province/ ; },
abstract = {This study focuses on improving the performance of steady-state visual evoked potential (SSVEP) in brain-computer interfaces (BCIs) for robotic control systems. The challenge lies in effectively reducing the impact of artifacts on raw data to enhance the performance both in quality and reliability. The proposed MVMD-MSI algorithm combines the advantages of multivariate variational mode decomposition (MVMD) and multivariate synchronization index (MSI). Compared to widely used algorithms, the novelty of this method is its capability of decomposing nonlinear and non-stationary EEG signals into intrinsic mode functions (IMF) across different frequency bands with the best center frequency and bandwidth. Therefore, SSVEP decoding performance can be improved by this method, and the effectiveness of MVMD-MSI is evaluated by the robot with 6 degrees-of-freedom. Offline experiments were conducted to optimize the algorithm's parameters, resulting in significant improvements. Additionally, the algorithm showed good performance even with fewer channels and shorter data lengths. In online experiments, the algorithm achieved an average accuracy of 98.31% at 1.8 s, confirming its feasibility and effectiveness for real-time SSVEP BCI-based robotic arm applications. The MVMD-MSI algorithm, as proposed, represents a significant advancement in SSVEP analysis for robotic control systems. It enhances decoding performance and shows promise for practical application in this field.},
}

@article {pmid39723643,
year = {2024},
author = {Tao, T and Liu, S and He, M and Zhao, M and Chen, C and Peng, J and Wang, Y and Cai, J and Xiong, J and Lai, C and Gu, W and Ying, M and Mao, J and Li, L and Jia, X and Wu, X and Peng, W and Zhang, X and Li, Y and Li, T and Wang, J and Shu, Q},
title = {Synchronous bilateral Wilms tumors are prone to develop independently and respond differently to preoperative chemotherapy.},
journal = {International journal of cancer},
volume = {},
number = {},
pages = {},
doi = {10.1002/ijc.35297},
pmid = {39723643},
issn = {1097-0215},
support = {32270853//National Natural Science Foundation of China/ ; U20A20137//National Natural Science Foundation of China/ ; LHDMZ23H160005//Joint Funds of the Zhejiang Provincial Natural Science Foundation of China/ ; 2024C03181//"Pioneer" and "Leading Goose" R&D Program of Zhejiang Province/ ; },
abstract = {Wilms tumor (WT) is the most common kidney cancer in infants and young children. The determination of the clonality of bilateral WTs is critical to the treatment, because lineage-independent and metastatic tumors may require different treatment strategies. Here we found synchronous bilateral WT (n = 24 tumors from 12 patients) responded differently to preoperative chemotherapy. Transcriptome, whole-exome and whole-genome analysis (n = 12 tumors from 6 patients) demonstrated that each side of bilateral WT was clonally independent in terms of somatic driver mutations, copy number variations and transcriptomic profile. Molecular timing analysis revealed distinct timing and patterns of chromosomal evolution and mutational processes between the two sides of WT. Mutations in WT1, CTNNB1 and copy-neutral loss of heterozygosity of 11p15.5 provide possible genetic predisposition for the early initiation of bilateral WT. Our results provide comprehensive evidence and new insights regarding the separate initiation and early embryonic development of bilateral WT, which may benefit clinical practices in treating metastatic or refractory bilateral WT.},
}

@article {pmid39720868,
year = {2024},
author = {Downey, JE and Schone, HR and Foldes, ST and Greenspon, C and Liu, F and Verbaarschot, C and Biro, D and Satzer, D and Moon, CH and Coffman, BA and Youssofzadeh, V and Fields, D and Hobbs, TG and Okorokova, E and Tyler-Kabara, EC and Warnke, PC and Gonzalez-Martinez, J and Hatsopoulos, NG and Bensmaia, SJ and Boninger, ML and Gaunt, RA and Collinger, JL},
title = {A Roadmap for Implanting Electrode Arrays to Evoke Tactile Sensations Through Intracortical Stimulation.},
journal = {Human brain mapping},
volume = {45},
number = {18},
pages = {e70118},
pmid = {39720868},
issn = {1097-0193},
support = {N66001-10-C-4056//Defense Advanced Research Projects Agency/ ; UH3 NS107714/NH/NIH HHS/United States ; },
mesh = {Humans ; Male ; Adult ; *Spinal Cord Injuries/physiopathology ; *Electrodes, Implanted ; Female ; *Somatosensory Cortex/physiology/diagnostic imaging ; *Brain-Computer Interfaces ; Middle Aged ; Touch Perception/physiology ; Electric Stimulation/methods ; Brain Mapping/methods ; Hand/physiology ; Magnetic Resonance Imaging ; },
abstract = {Intracortical microstimulation (ICMS) is a method for restoring sensation to people with paralysis as part of a bidirectional brain-computer interface (BCI) to restore upper limb function. Evoking tactile sensations of the hand through ICMS requires precise targeting of implanted electrodes. Here we describe the presurgical imaging procedures used to generate functional maps of the hand area of the somatosensory cortex and subsequent planning that guided the implantation of intracortical microelectrode arrays. In five participants with cervical spinal cord injury, across two study locations, this procedure successfully enabled ICMS-evoked sensations localized to at least the first four digits of the hand. The imaging and planning procedures developed through this clinical trial provide a roadmap for other BCI studies to ensure the successful placement of stimulation electrodes.},
}

Humans
Male
Adult
*Spinal Cord Injuries/physiopathology
*Electrodes, Implanted
Female
*Somatosensory Cortex/physiology/diagnostic imaging
*Brain-Computer Interfaces
Middle Aged
Touch Perception/physiology
Electric Stimulation/methods
Brain Mapping/methods
Hand/physiology
Magnetic Resonance Imaging

@article {pmid39720668,
year = {2024},
author = {Li, X and Chu, Y and Wu, X},
title = {3D convolutional neural network based on spatial-spectral feature pictures learning for decoding motor imagery EEG signal.},
journal = {Frontiers in neurorobotics},
volume = {18},
number = {},
pages = {1485640},
pmid = {39720668},
issn = {1662-5218},
abstract = {Non-invasive brain-computer interfaces (BCI) hold great promise in the field of neurorehabilitation. They are easy to use and do not require surgery, particularly in the area of motor imagery electroencephalography (EEG). However, motor imagery EEG signals often have a low signal-to-noise ratio and limited spatial and temporal resolution. Traditional deep neural networks typically only focus on the spatial and temporal features of EEG, resulting in relatively low decoding and accuracy rates for motor imagery tasks. To address these challenges, this paper proposes a 3D Convolutional Neural Network (P-3DCNN) decoding method that jointly learns spatial-frequency feature maps from the frequency and spatial domains of the EEG signals. First, the Welch method is used to calculate the frequency band power spectrum of the EEG, and a 2D matrix representing the spatial topology distribution of the electrodes is constructed. These spatial-frequency representations are then generated through cubic interpolation of the temporal EEG data. Next, the paper designs a 3DCNN network with 1D and 2D convolutional layers in series to optimize the convolutional kernel parameters and effectively learn the spatial-frequency features of the EEG. Batch normalization and dropout are also applied to improve the training speed and classification performance of the network. Finally, through experiments, the proposed method is compared to various classic machine learning and deep learning techniques. The results show an average decoding accuracy rate of 86.69%, surpassing other advanced networks. This demonstrates the effectiveness of our approach in decoding motor imagery EEG and offers valuable insights for the development of BCI.},
}

@article {pmid39720230,
year = {2024},
author = {Hu, F and He, K and Qian, M and Liu, X and Qiao, Z and Zhang, L and Xiong, J},
title = {STAFNet: an adaptive multi-feature learning network via spatiotemporal fusion for EEG-based emotion recognition.},
journal = {Frontiers in neuroscience},
volume = {18},
number = {},
pages = {1519970},
pmid = {39720230},
issn = {1662-4548},
abstract = {INTRODUCTION: Emotion recognition using electroencephalography (EEG) is a key aspect of brain-computer interface research. Achieving precision requires effectively extracting and integrating both spatial and temporal features. However, many studies focus on a single dimension, neglecting the interplay and complementarity of multi-feature information, and the importance of fully integrating spatial and temporal dynamics to enhance performance.

METHODS: We propose the Spatiotemporal Adaptive Fusion Network (STAFNet), a novel framework combining adaptive graph convolution and temporal transformers to enhance the accuracy and robustness of EEG-based emotion recognition. The model includes an adaptive graph convolutional module to capture brain connectivity patterns through spatial dynamic evolution and a multi-structured transformer fusion module to integrate latent correlations between spatial and temporal features for emotion classification.

RESULTS: Extensive experiments were conducted on the SEED and SEED-IV datasets to evaluate the performance of STAFNet. The model achieved accuracies of 97.89% and 93.64%, respectively, outperforming state-of-the-art methods. Interpretability analyses, including confusion matrices and t-SNE visualizations, were employed to examine the influence of different emotions on the model's recognition performance. Furthermore, an investigation of varying GCN layer depths demonstrated that STAFNet effectively mitigates the over-smoothing issue in deeper GCN architectures.

DISCUSSION: In summary, the findings validate the effectiveness of STAFNet in EEG-based emotion recognition. The results emphasize the critical role of spatiotemporal feature extraction and introduce an innovative framework for feature fusion, advancing the state of the art in emotion recognition.},
}

@article {pmid39719191,
year = {2024},
author = {Rudroff, T},
title = {Decoding thoughts, encoding ethics: A narrative review of the BCI-AI revolution.},
journal = {Brain research},
volume = {1850},
number = {},
pages = {149423},
doi = {10.1016/j.brainres.2024.149423},
pmid = {39719191},
issn = {1872-6240},
abstract = {OBJECTIVES: This narrative review aims to analyze mechanisms underlying Brain-Computer Interface (BCI) and Artificial Intelligence (AI) integration, evaluate recent advances in signal acquisition and processing techniques, and assess AI-enhanced neural decoding strategies. The review identifies critical research gaps and examines emerging solutions across multiple domains of BCI-AI integration.

METHODS: A narrative review was conducted using major biomedical and scientific databases including PubMed, Web of Science, IEEE Xplore, and Scopus (2014-2024). Literature was analyzed to identify key developments in BCI-AI integration, with particular emphasis on recent advances (2019-2024). The review process involved thematic analysis of selected publications focusing on practical applications, technical innovations, and emerging challenges.

RESULTS: Recent advances demonstrate significant improvements in BCI-AI systems: 1) High-density electrode arrays achieve spatial resolution up to 5 mm, with stable recordings over 15 months; 2) Deep learning decoders show 40 % improvement in information transfer rates compared to traditional methods; 3) Adaptive algorithms maintain >90 % success rates in motor control tasks over 200-day periods without recalibration; 4) Novel closed-loop optimization frameworks reduce user training time by 55 % while improving accuracy. Latest developments in flexible neural interfaces and self-supervised learning approaches show promise in addressing long-term stability and cross-user generalization challenges.

CONCLUSIONS: BCI-AI integration shows remarkable progress in improving signal quality, decoding accuracy, and user adaptation. While challenges remain in long-term stability and user training, advances in adaptive algorithms and feedback mechanisms demonstrate the technology's growing viability for clinical applications. Recent innovations in electrode technology, AI architectures, and closed-loop systems, combined with emerging standardization frameworks, suggest accelerating progress toward widespread therapeutic use and human augmentation applications.},
}

@article {pmid39719121,
year = {2025},
author = {Li, K and Chen, P and Chen, Q and Li, X},
title = {A hybrid network using transformer with modified locally linear embedding and sliding window convolution for EEG decoding.},
journal = {Journal of neural engineering},
volume = {21},
number = {6},
pages = {},
doi = {10.1088/1741-2552/ada30b},
pmid = {39719121},
issn = {1741-2552},
mesh = {*Electroencephalography/methods ; *Brain-Computer Interfaces ; Humans ; *Neural Networks, Computer ; Imagination/physiology ; Algorithms ; Linear Models ; Signal Processing, Computer-Assisted ; },
abstract = {Objective. Brain-computer interface(BCI) is leveraged by artificial intelligence in EEG signal decoding, which makes it possible to become a new means of human-machine interaction. However, the performance of current EEG decoding methods is still insufficient for clinical applications because of inadequate EEG information extraction and limited computational resources in hospitals. This paper introduces a hybrid network that employs a transformer with modified locally linear embedding and sliding window convolution for EEG decoding.Approach. This network separately extracts channel and temporal features from EEG signals, subsequently fusing these features using a cross-attention mechanism. Simultaneously, manifold learning is employed to lower the computational burden of the model by mapping the high-dimensional EEG data to a low-dimensional space by its dimension reduction function.Main results. The proposed model achieves accuracy rates of 84.44%, 94.96%, and 82.79% on the BCI Competition IV dataset 2a, high gamma dataset, and a self-constructed motor imagery (MI) dataset from the left and right hand fist-clenching tests respectively. The results indicate our model outperforms the baseline models by EEG-channel transformer with dimension-reduced EEG data and window attention with sliding window convolution. Additionally, to enhance the interpretability of the model, features preceding the temporal feature extraction network were visualized. This visualization promotes the understanding of how the model prefers task-related channels.Significance. The transformer-based method makes the MI-EEG decoding more practical for further clinical applications.},
}

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

@article {pmid39718409,
year = {2024},
author = {Ma, X and Chen, W and Pei, Z and Zhang, J},
title = {An attention-based motor imagery brain-computer interface system for lower limb exoskeletons.},
journal = {The Review of scientific instruments},
volume = {95},
number = {12},
pages = {},
doi = {10.1063/5.0243337},
pmid = {39718409},
issn = {1089-7623},
mesh = {*Brain-Computer Interfaces ; Humans ; *Lower Extremity/physiology ; *Exoskeleton Device ; *Electroencephalography/instrumentation ; Attention/physiology ; Neural Networks, Computer ; Adult ; Male ; Imagination/physiology ; },
abstract = {Lower-limb exoskeletons have become increasingly popular in rehabilitation to help patients with disabilities regain mobility and independence. Brain-computer interface (BCI) offers a natural control method for these exoskeletons, allowing users to operate them through their electroencephalogram (EEG) signals. However, the limited EEG decoding performance of the BCI system restricts its application for lower limb exoskeletons. To address this challenge, we propose an attention-based motor imagery BCI system for lower limb exoskeletons. The decoding module of the proposed BCI system combines the convolutional neural network (CNN) with a lightweight attention module. The CNN aims to extract meaningful features from EEG signals, while the lightweight attention module aims to capture global dependencies among these features. The experiments are divided into offline and online experiments. The offline experiment is conducted to evaluate the effectiveness of different decoding methods, while the online experiment is conducted on a customized lower limb exoskeleton to evaluate the proposed BCI system. Eight subjects are recruited for the experiments. The experimental results demonstrate the great classification performance of the decoding method and validate the feasibility of the proposed BCI system. Our approach establishes a promising BCI system for the lower limb exoskeleton and is expected to achieve a more effective and user-friendly rehabilitation process.},
}

*Brain-Computer Interfaces
Humans
*Lower Extremity/physiology
*Exoskeleton Device
*Electroencephalography/instrumentation
Attention/physiology
Neural Networks, Computer
Adult
Male
Imagination/physiology

@article {pmid39715900,
year = {2024},
author = {Robinson, JT and Norman, SL and Angle, MR and Constandinou, TG and Denison, T and Donoghue, JP and Field, RM and Forsland, A and Kouider, S and Millán, JDR and Michaels, JA and Orsborn, AL and Pandarinath, C and Pruszynski, JA and Rozell, CJ and Shah, NP and Shanechi, MM and Shoaran, M and Sheth, SA and Stavisky, SD and Trautmann, E and Vachicouras, N and Xie, C},
title = {An application-based taxonomy for brain-computer interfaces.},
journal = {Nature biomedical engineering},
volume = {},
number = {},
pages = {},
pmid = {39715900},
issn = {2157-846X},
}

@article {pmid39712143,
year = {2024},
author = {Bai, G and Jin, J and Xu, R and Wang, X and Cichocki, A},
title = {A novel dual-step transfer framework based on domain selection and feature alignment for motor imagery decoding.},
journal = {Cognitive neurodynamics},
volume = {18},
number = {6},
pages = {3549-3563},
pmid = {39712143},
issn = {1871-4080},
abstract = {In brain-computer interfaces (BCIs) based on motor imagery (MI), reducing calibration time is gradually becoming an urgent issue in practical applications. Recently, transfer learning (TL) has demonstrated its effectiveness in reducing calibration time in MI-BCI. However, the different data distribution of subjects greatly affects the application effect of TL in MI-BCI. Therefore, this paper combines data alignment, source domain selection, and feature alignment into the MI-TL. We propose a novel dual-step transfer framework based on source domain selection and feature alignment. First, the source and target domains are aligned using a pre-calibration strategy (PS), and then a sequential reverse selection method is proposed to match the optimal source domain for each target domain with the designed dual model selection strategy. We use filter bank regularization common space pattern (FBRCSP) to obtain more features and introduce manifold embedded distribution alignment (MEDA) to correct the prediction results of the support vector machine (SVM). The experimental results on two competition public datasets (BCI competition IV Dataset 1 and Dataset 2a) and our dataset show that the average classification accuracy of the proposed framework is higher than the baseline method (no domain selection and no feature alignment), which reaches 84.12%, 79.91%, and 78.45%, respectively. And the computational cost is reduced by half compared with the baseline method.},
}

@article {pmid39712134,
year = {2024},
author = {Zhang, J and Shen, C and Chen, W and Ma, X and Liang, Z and Zhang, Y},
title = {Decoding of movement-related cortical potentials at different speeds.},
journal = {Cognitive neurodynamics},
volume = {18},
number = {6},
pages = {3859-3872},
pmid = {39712134},
issn = {1871-4080},
abstract = {The decoding of electroencephalogram (EEG) signals, especially motion-related cortical potentials (MRCP), is vital for the early detection of motor intent before movement execution. To enhance the decoding accuracy of MRCP and promote the application of early motion intention in active rehabilitation training, we propose a method for decoding MRCP signals. Specifically, an experimental paradigm is designed for the efficient capture of MRCP signals. Moreover, a feature extraction method based on differentiation is proposed to effectively characterize action variability. Six subjects were recruited to validate the effectiveness of the decoding method. Experiments such as fixed-window classification, sliding-window detection, and asynchronous analysis demonstrate that the method can detect motion intention 316 milliseconds before action execution and is capable of continuously detecting both rapid and slow movements.},
}

@article {pmid39712133,
year = {2024},
author = {Wu, C and Wang, Y and Qiu, S and He, H},
title = {A bimodal deep learning network based on CNN for fine motor imagery.},
journal = {Cognitive neurodynamics},
volume = {18},
number = {6},
pages = {3791-3804},
pmid = {39712133},
issn = {1871-4080},
abstract = {Motor imagery (MI) is an important brain-computer interface (BCI) paradigm. The traditional MI paradigm (imagining different limbs) limits the intuitive control of the outer devices, while fine MI paradigm (imagining different joint movements from the same limb) can control the mechanical arm without cognitive disconnection. However, the decoding performance of fine MI limits its application. Electroencephalogram (EEG) and functional near-infrared spectroscopy (fNIRS) are widely used in BCI systems because of their portability and easy operation. In this study, a fine MI paradigm including four classes (hand, wrist, shoulder and rest) was designed, and the data of EEG-fNIRS bimodal brain activity was collected from 12 subjects. Event-related desynchronization (ERD) from EEG signals shows a contralateral dominant phenomenon, and there is difference between the ERD of the four classes. For fNIRS signal in the time dimension, the time periods with significant difference can be observed in the activation patterns of four MI tasks. Spatially, the signal peak based brain topographic map also shows difference of these four MI tasks. The EEG signal and fNIRS signal of these four classes are distinguishable. In this study, a bimodal fusion network is proposed to improve the fine MI tasks decoding performance. The features of these two modalities are extracted separately by two feature extractors based on convolutional neural networks (CNN). The recognition performance was significantly improved by the bimodal method proposed in this study, compared with the performance of the single-modal network. The proposed method outperformed all comparison methods, and achieved a four-class accuracy of 58.96%. This paper demonstrates the feasibility of EEG and fNIRS bimodal BCI systems for fine MI, and shows the effectiveness of the proposed bimodal fusion method. This research is supposed to support fine MI-based BCI systems with theories and techniques.},
}

@article {pmid39712131,
year = {2024},
author = {Meng, M and Xu, B and Ma, Y and Gao, Y and Luo, Z},
title = {STGAT-CS: spatio-temporal-graph attention network based channel selection for MI-based BCI.},
journal = {Cognitive neurodynamics},
volume = {18},
number = {6},
pages = {3663-3678},
pmid = {39712131},
issn = {1871-4080},
abstract = {Brain-computer interface (BCI) based on the motor imagery paradigm typically utilizes multi-channel electroencephalogram (EEG) to ensure accurate capture of physiological phenomena. However, excessive channels often contain redundant information and noise, which can significantly degrade BCI performance. Although there have been numerous studies on EEG channel selection, most of them require manual feature extraction, and the extracted features are difficult to fully represent the effective information of EEG signals. In this paper, we propose a spatio-temporal-graph attention network for channel selection (STGAT-CS) of EEG signals. We consider the EEG channels and their inter-channel connectivity as a graph and treat the channel selection problem as a node classification problem on the graph. We leverage the multi-head attention mechanism of graph attention network to dynamically capture topological relationships between nodes and update node features accordingly. Additionally, we introduce one-dimensional convolution to automatically extract temporal features from each channel in the original EEG signal, thereby obtaining more comprehensive spatiotemporal characteristics. In the classification tasks of the BCI Competition III Dataset IVa and BCI Competition IV Dataset I, STGAT-CS achieved average accuracies of 91.5% and 85.4% respectively, demonstrating the effectiveness of the proposed method.},
}

@article {pmid39712122,
year = {2024},
author = {Li, M and Li, J and Zheng, X and Ge, J and Xu, G},
title = {MSHANet: a multi-scale residual network with hybrid attention for motor imagery EEG decoding.},
journal = {Cognitive neurodynamics},
volume = {18},
number = {6},
pages = {3463-3476},
pmid = {39712122},
issn = {1871-4080},
abstract = {EEG decoding plays a crucial role in the development of motor imagery brain-computer interface. Deep learning has great potential to automatically extract EEG features for end-to-end decoding. Currently, the deep learning is faced with the chanllenge of decoding from a large amount of time-variant EEG to retain a stable peroformance with different sessions. This study proposes a multi-scale residual network with hybrid attention (MSHANet) to decode four motor imagery classes. The MSHANet combines a multi-head attention and squeeze-and-excitation attention to hybridly focus on important information of the EEG features; and applies a multi-scale residual block to extracts rich EEG features, sharing part of the block parameters to extract common features. Compared with seven state-of-the-art methods, the MSHANet exhits the best accuracy on BCI Competition IV 2a with an accuracy of 83.18% for session- specific task and 80.09% for cross-session task. Thus, the proposed MSHANet decodes the time-varying EEG robustly and can save the time cost of MI-BCI, which is beneficial for long-term use.},
}

@article {pmid39712121,
year = {2024},
author = {Rahman, N and Khan, DM and Masroor, K and Arshad, M and Rafiq, A and Fahim, SM},
title = {Advances in brain-computer interface for decoding speech imagery from EEG signals: a systematic review.},
journal = {Cognitive neurodynamics},
volume = {18},
number = {6},
pages = {3565-3583},
pmid = {39712121},
issn = {1871-4080},
abstract = {Numerous individuals encounter challenges in verbal communication due to various factors, including physical disabilities, neurological disorders, and strokes. In response to this pressing need, technology has actively pursued solutions to bridge the communication gap, recognizing the inherent difficulties faced in verbal communication, particularly in contexts where traditional methods may be inadequate. Electroencephalogram (EEG) has emerged as a primary non-invasive method for measuring brain activity, offering valuable insights from a cognitive neurodevelopmental perspective. It forms the basis for Brain-Computer Interfaces (BCIs) that provide a communication channel for individuals with neurological impairments, thereby empowering them to express themselves effectively. EEG-based BCIs, especially those adapted to decode imagined speech from EEG signals, represent a significant advancement in enabling individuals with speech disabilities to communicate through text or synthesized speech. By utilizing cognitive neurodevelopmental insights, researchers have been able to develop innovative approaches for interpreting EEG signals and translating them into meaningful communication outputs. To aid researchers in effectively addressing this complex challenge, this review article synthesizes key findings from state-of-the-art significant studies. It investigates into the methodologies employed by various researchers, including preprocessing techniques, feature extraction methods, and classification algorithms utilizing Deep Learning and Machine Learning approaches and their integration. Furthermore, the review outlines the potential avenues for future research, with the goal of advancing the practical implementation of EEG-based BCI systems for decoding imagined speech from a cognitive neurodevelopmental perspective.},
}

@article {pmid39712116,
year = {2024},
author = {Tang, C and Gao, T and Wang, G and Chen, B},
title = {Coherence-based channel selection and Riemannian geometry features for magnetoencephalography decoding.},
journal = {Cognitive neurodynamics},
volume = {18},
number = {6},
pages = {3535-3548},
pmid = {39712116},
issn = {1871-4080},
abstract = {Magnetoencephalography (MEG) records the extremely weak magnetic fields on the surface of the scalp through highly sensitive sensors. Multi-channel MEG data provide higher spatial and temporal resolution when measuring brain activities, and can be applied for brain-computer interfaces as well. However, a large number of channels leads to high computational complexity and can potentially impact decoding accuracy. To improve the accuracy of MEG decoding, this paper proposes a new coherence-based channel selection method that effectively identifies task-relevant channels, reducing the presence of noisy and redundant information. Riemannian geometry is then used to extract effective features from selected channels of MEG data. Finally, MEG decoding is achieved by training a support vector machine classifier with the Radial Basis Function kernel. Experiments were conducted on two public MEG datasets to validate the effectiveness of the proposed method. The results from Dataset 1 show that Riemannian geometry achieves higher classification accuracy (compared to common spatial patterns and power spectral density) in the single-subject visual decoding task. Moreover, coherence-based channel selection significantly (P = 0.0002) outperforms the use of all channels. Moving on to Dataset 2, the results reveal that coherence-based channel selection is also significantly (P <0.0001) superior to all channels and channels around C3 and C4 in cross-session mental imagery decoding tasks. Additionally, the proposed method outperforms state-of-the-art methods in motor imagery tasks.},
}

@article {pmid39712104,
year = {2024},
author = {Yu, H and Hu, Z and Zhao, Q and Liu, J},
title = {Deep source transfer learning for the estimation of internal brain dynamics using scalp EEG.},
journal = {Cognitive neurodynamics},
volume = {18},
number = {6},
pages = {3507-3520},
pmid = {39712104},
issn = {1871-4080},
abstract = {Electroencephalography (EEG) provides high temporal resolution neural data for brain-computer interfacing via noninvasive electrophysiological recording. Estimating the internal brain activity by means of source imaging techniques can further improve the spatial resolution of EEG and enhance the reliability of neural decoding and brain-computer interaction. In this work, we propose a novel EEG data-driven source imaging scheme for precise and efficient estimation of macroscale spatiotemporal brain dynamics across thalamus and cortical regions with deep learning methods. A deep source imaging framework with a convolutional-recurrent neural network is designed to estimate the internal brain dynamics from high-density EEG recordings. Moreover, a brain model including 210 cortical regions and 16 thalamic nuclei is established based on human brain connectome to provide synthetic training data, which manifests intrinsic characteristics of underlying brain dynamics in spontaneous, stimulation-evoked, and pathological states. Transfer learning algorithm is further applied to the trained network to reduce the dynamical differences between synthetic and realistic EEG. Extensive experiments exhibit that the proposed deep-learning method can accurately estimate the spatial and temporal activity of brain sources and achieves superior performance compared to the state-of-the-art approaches. Moreover, the EEG data-driven source imaging framework is effective in the location of seizure onset zone in epilepsy and reconstruction of dynamical thalamocortical interactions during sensory processing of acupuncture stimulation, implying its applicability in brain-computer interfacing for neuroscience research and clinical applications.},
}

@article {pmid39712096,
year = {2024},
author = {Zhang, Z and Chen, Y and Zhao, X and Fan, W and Peng, D and Li, T and Zhao, L and Fu, Y},
title = {A review of ethical considerations for the medical applications of brain-computer interfaces.},
journal = {Cognitive neurodynamics},
volume = {18},
number = {6},
pages = {3603-3614},
pmid = {39712096},
issn = {1871-4080},
abstract = {The development and potential applications of brain-computer interfaces (BCIs) are directly related to the human brain and may have adverse effects on the users' physical and mental health. Ethical issues, particularly those associated with BCIs, including both non-medical and medical applications, have captured societal attention. This article initially reviews the application of three ethical frameworks in BCI technology: consequentialism, deontology, and virtue ethics. Subsequently, it introduces the ethical standards under consideration within the medical objective framework for BCI medical applications. Finally, the paper discusses and forecasts the ethical standards for BCI medical applications. The paper emphasizes the necessity to differentiate between the ethical issues of implantable and non-implantable BCIs, to approach the research on BCI-based "controlling the brain" with caution, and to establish standardized operational procedures and efficacy evaluation methods for BCI medical applications. This paper aims to provide ideas for the establishment of ethical standards in BCI medical applications.},
}

@article {pmid39712090,
year = {2024},
author = {Huang, Y and Huan, Y and Zou, Z and Wang, Y and Gao, X and Zheng, L},
title = {Data-driven natural computational psychophysiology in class.},
journal = {Cognitive neurodynamics},
volume = {18},
number = {6},
pages = {3477-3489},
pmid = {39712090},
issn = {1871-4080},
abstract = {Objective. The assessment of mental fatigue (MF) and attention span in educational and healthcare settings frequently relies on subjective scales or methods such as induced-task interruption tools. However, these approaches are deficient in real-time evaluation and dynamic definitions. To address this gap, this paper proposes a Continuous Quantitative Scale (CQS) that allows for the natural and real-time measurement of MF based on group-synchronized electroencephalogram (EEG) data. Approach. In this study, computational psychophysiology was used to measure MF scores during a realistic class. Our methodology continuously monitored participants' psychological states without interrupting their regular routines, providing an objective evaluation. By analyzing multi-subject brain-computer interface (mBCI) data with a collaborative computing approach, the group-synchronized data were obtained from 10 healthy participants to assess MF levels. Each participant wore an EEG headset for only 10 min of preparation before performing a sustained task for 80 min. Main results. Our findings indicate that a lecture duration of 18.9 min is most effective, while a duration of 43.1 min leads to heightened MF levels. By focusing on the group-level simultaneous data analysis, the effects of individual variability were mitigated and the efficiency of cognitive computing was improved. From the perspective of a neurocomputational measure, these results confirm previous research. Significance. The proposed CQS provides a reliable, objective, memory- and emotion-free approach to the assessment of MF and attention span. These findings have significant implications not only for education, but also for the study of group cognitive mechanisms and for improving the quality of mental healthcare.},
}

@article {pmid39711742,
year = {2024},
author = {Li, J and Nan, Z and Qi, G and Cai, J and Zhao, X and Li, X and Liu, S and Wang, Y and Wu, Y and Miao, X and Yu, G},
title = {Assessing severity of pediatric pneumonia using multimodal transformers with multi-task learning.},
journal = {Digital health},
volume = {10},
number = {},
pages = {20552076241305168},
pmid = {39711742},
issn = {2055-2076},
abstract = {OBJECTIVE: While current multimodal approaches in the diagnosis and severity assessment of pneumonia demonstrate remarkable performance, they frequently overlook the issue of modality absence-a common challenge in clinical practice. Thus, we present the robust multimodal transformer (RMT) model, crafted to bridge this gap. The RMT model aims to enhance diagnosis and severity assessment accuracy in situations with incomplete data, thereby ensuring it meets the complex needs of real-world clinical settings.

METHOD: The RMT model leverages multimodal data, integrating X-ray images and clinical text data through a sophisticated AI-driven framework. It employs a Transformer-based architecture, enhanced by multi-task learning and mask attention mechanism. This approach aims to optimize the model's performance across different modalities, particularly under conditions of modality absence.

RESULTS: The RMT model demonstrates superior performance over traditional diagnostic methods and baseline models in accuracy, precision, sensitivity, and specificity. In tests involving various scenarios, including single-modal and multimodal tasks, the model shows remarkable robustness in handling incomplete data. Its effectiveness is further validated through extensive comparative analysis and ablation studies.

CONCLUSION: The RMT model represents a substantial advancement in pediatric pneumonia severity assessment. It successfully harnesses multimodal data and advanced AI techniques to improve assessment precision. While the RMT model sets a new precedent in AI applications in medical diagnostics, the development of a comprehensive pediatric pneumonia dataset marks a pivotal contribution, providing a robust foundation for future research.},
}