@article {pmid42913687,
year = {2026},
author = {Kokorin, K and Mu, J and Grayden, DB and John, SE and Yoo, PE},
title = {Decoding attempted handwritten characters based on temporal endovascular neural features.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/aeb82b},
pmid = {42913687},
issn = {1741-2552},
abstract = {We evaluated temporal neural features for motor decoding using a minimally-invasive endovascular brain-computer interfaces (BCI). We hypothesised that relatively long compound movements generate distinct patterns of neural activity that are encoded by temporal features. Approach. A single participant implanted with the Stentrode endovascular BCI attempted to hand write English characters, which were then classified offline and online based on the recorded brain activity. The input to the model was a one-dimensional trajectory of high-gamma power, summed across all channels to isolate the effect of temporal information on decoding performance. Main results. A small number of classes across multiple months was reliably decoded, showing that temporal decoding could identify stable patterns of brain activity associated with each handwriting movement. Across four online sessions, where each session contained 20-30 trials/character and 80-180 rest trials, the best performing (3-class) model achieved a mean rest recall of 87% as well as character recall and precision of 65% and 59%, respectively. These results showed that attempted handwriting can be used in an online multi-class selection BCI and that temporal neural features capture the internal user context. Linear warping of trials to minimise temporal variability, where all trials were modified to have the same duration, led to significantly increased decoding performance compared to classifying raw trials. Significance. The results from our participant showed that decoding brain activity based on the differences between feature trajectories across time can be an effective strategy for discriminating different movements. Therefore, BCIs may be able to use temporal information to reduce their spatial footprint, likely making them less invasive. Overall, our study shows that further investigation of temporal decoding across multiple participants and other recording modalities is warranted. Strategies for automatic temporal alignment are likely to further improve the performance of these systems.},
}

@article {pmid42913696,
year = {2026},
author = {Hart, DM and Polichtchouk, S and Dannhauer, M},
title = {Transformer networks are robust learners of source localization for EEG measurements.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/aeb82c},
pmid = {42913696},
issn = {1741-2552},
abstract = {EEG source localization is an ill-posed inverse problem complicated by measurement noise, electrode dropout from sensor miscalibration, and variability in the number of active sources. Classical methods such as eLORETA and LCMV beamforming impose regularization assumptions that degrade under realistic noise conditions and do not naturally extend to multi-source scenarios. Recent neural network approaches improve localization accuracy but are typically trained and evaluated under idealized conditions that limit their applicability to real-world EEG acquisition. This work aims to develop and evaluate a source localization model that generalizes across the noise levels, source counts, and electrode configurations encountered in clinical and experimental EEG settings. Approach: We present a transformer-based model for single- and multi-source EEG localization. Hundreds of thousands of simulated EEG measurements are generated from known source locations, spanning a range of noise levels, source counts, and electrode dropout configurations, to train and evaluate the model. We establish a systematic evaluation framework that explicitly factors in source region density, signal-to-noise ratio, and electrode availability, and benchmark the transformer against classical (eLORETA, LCMV beamforming) and deep learning baselines across these conditions. We additionally conduct a thorough ablation study on network architecture and training pipeline design. Main results: Across all tested conditions, the transformer consistently outperforms classical and deep learning approaches, with improvements as high as 12\,mm in mean localization error (MLE) for multi-source scenarios under high noise. Additionally, when number of sources is unknown, the transformer can correctly predict the number and location of sources and drastically out perform the threshold-based approach (e.g an F1 Score of 68\% vs 12\% when simulated at 25\% noise). The multiembedding transformers maintains reasonable performance under conditions such as electrode dropout and increased noise. Significance: By explicitly modeling the noise, electrode dropout, and multi-source variability that characterize real EEG acquisition, this work addresses a key gap between benchmark performance and experimental utility in EEG source localization. These results lay the groundwork for more robust localization techniques applicable to real EEG data, with potential relevance to applications such as brain-computer interfaces. The data and code for this work will be made publicly available to support further development and benchmarking in the field.},
}

@article {pmid42909910,
year = {2026},
author = {Zheng, Q and Wen, J and Yuan, W and Qin, J and Duanmu, X and Wu, H and Zhu, Z and Guo, T and Wu, C and Zhou, C and Wu, J and Chen, J and Fang, Y and Zhu, B and Yan, Y and Tian, J and Zhang, B and He, H and Zhang, M and Guan, X and Xu, X},
title = {A clinically interpretable model for Parkinson's disease: insights from complementary MRI modalities.},
journal = {Brain communications},
volume = {8},
number = {5},
pages = {fcag383},
pmid = {42909910},
issn = {2632-1297},
abstract = {Accurate identification of Parkinson's disease may remain challenging in certain clinical scenarios, highlighting the need for a clinical decision-support model based on complementary MRI modalities. The case-control study included 462 participants, split into training [247 Parkinson's disease, 122 normal controls (NC)] and testing (60 Parkinson's disease, 33 NC) cohorts to construct and validate a clinical decision-support model. We employed least absolute shrinkage and selection operator and logistic regression analysis to develop the model utilizing quantitative susceptibility mapping, arterial spin labelling and T1-weighted images. The model was assessed for calibration, discrimination and clinical utility, with individual predicted scores correlated to Parkinson's disease clinical symptoms. The model's performance was also evaluated independently in multiple system atrophy, progressive supranuclear palsy and essential tremor, with DeLong test comparisons. The model was trained in the training cohort (Parkinson's disease: 61.39 ± 8.06 years old, 134 males; NC: 61.29 ± 7.56 years old, 54 males) with an area under the ROC curve (AUC) of 0.842 [95% confidence interval (CI) 0.798-0.883] and 0.771 (95% CI 0.657-0.870) in the testing cohort, whose clinical utility was confirmed. The predicted scores were positively correlated with Unified Parkinson's Disease Rating Scale II (P = 0.034) and III (P = 0.015), while negatively correlated with the Purdue Pegboard total (P = 0.010) and assembly scores (P = 0.007). When discriminating each disease group from NC, the model's performance to identify multiple system atrophy and progressive supranuclear palsy with AUC values of 0.881 and 0.895 was superior than essential tremor with an AUC of 0.709 before false discovery rate correction (multiple system atrophy-NC versus essential tremor-NC: P = 0.041; progressive supranuclear palsy-NC versus essential tremor-NC: P = 0.010). The model based on complementary MRI modalities was an objective clinical decision-support measure for Parkinson's disease with potential applications in other parkinsonisms (multiple system atrophy, progressive supranuclear palsy).},
}

@article {pmid42910633,
year = {2026},
author = {Cai, X and Liu, Y and Sheng, X},
title = {Engineering Bidirectional Neuromodulation: A Review of Neural Excitation and Inhibition Technologies.},
journal = {Advanced materials (Deerfield Beach, Fla.)},
volume = {},
number = {},
pages = {},
doi = {10.1002/adma.75480},
pmid = {42910633},
issn = {1521-4095},
support = {YIRG MOH-001243-00//NMRC Open Fund-Young Individual Research Grant/ ; NRF2021-NRF-ANR004//NRF-ANR Joint Grant Call/ ; 22-4974-A0003//National University of Singapore/ ; 2024Z02ORD001//Tsinghua University/ ; T2425003//National Natural Science Foundation of China/ ; 52272277//National Natural Science Foundation of China/ ; 20253930040//Tsinghua-Toyota Joint Research Fund/ ; },
abstract = {Neuromodulation provides a means to perturb neural activity for investigating nervous system function and treating neurological disorders. Advances in stimulation strategies, molecular tools, functional materials and neural interfaces have progressively expanded the precision and versatility of such interventions. Yet neural activity can require regulation in opposing directions across different cellular, circuit and pathological states, motivating bidirectional neuromodulation that can both enhance and suppress neural activity. In this review, we examine the mechanisms and technologies that enable such bidirectional control across electrical, optical, thermal, mechanical, magnetic and chemical modalities. We compare these approaches in terms of spatiotemporal performance, penetration depth, specificity and safety, and examine the materials, transduction strategies and device architectures used to implement bidirectional neural control. We further discuss the applications of bidirectional neuromodulation in neuroscience research and the treatment of neurological disorders. Finally, we discuss the technical challenges and future directions toward more precise, independent and integrated bidirectional control, including its extension to closed-loop adaptive regulation.},
}

@article {pmid42911760,
year = {2026},
author = {Zhao, G and Feng, Z and Wang, Z},
title = {Adaptive local temporal weighting for short-window EEG-based SSVEP recognition using FB-eTRCA.},
journal = {Science progress},
volume = {109},
number = {4},
pages = {368504261501367},
doi = {10.1177/00368504261501367},
pmid = {42911760},
issn = {2047-7163},
mesh = {Humans ; *Electroencephalography/methods ; *Evoked Potentials, Visual/physiology ; Algorithms ; Signal Processing, Computer-Assisted ; },
abstract = {To improve short-window SSVEP recognition, we integrated adaptive temporal local weighting (ATLW) with filter-bank ensemble task-related component analysis (FB-eTRCA). ATLW estimates local discriminability from training trials, converts window scores into sample weights, and applies the training-derived weights to training and test trials within each cross-validation fold. Experiments were conducted on two public 40-target SSVEP datasets: the Benchmark dataset containing 35 subjects and six blocks, and the BETA database containing 70 subjects and four blocks. Seven observation windows ranging from 0.2 to 1.4 s were evaluated using subject-specific leave-one-block-out validation, with six folds for the Benchmark dataset and four folds for the BETA database. On the Benchmark dataset, only the 0.2-s accuracy gain remained significant after Holm correction (44.79% vs 44.07%; adjusted p = 0.0300); other windows were nonsignificant. On the Benchmark dataset, ablation analysis showed that adaptive weighting was more stable than fixed temporal weighting. After Holm correction, eTRCA+ATLW significantly outperformed eTRCA at 0.4 s (gain: 0.86 percentage points; adjusted p = 0.0023), but not at 0.2 s, supporting transferability to another TRCA-based decoder. EEGNet and BiLSTM were included as representative CNN and RNN baselines; both performed below the TRCA-family methods under the present subject-specific limited-calibration protocol. On the BETA database, FB-eTRCA+ATLW remained close to FB-eTRCA across all seven observation windows without BETA-specific parameter retuning, but no paired difference was statistically significant. These findings indicate that ATLW is a lightweight temporal refinement that can be applied to more than one TRCA-based decoder. Its measurable benefit may depend on SSVEP signal quality and the amount of available subject-specific calibration data. Accuracy increased with longer windows, whereas the information transfer rate peaked at 0.6 s on the Benchmark dataset. Therefore, no observation window was optimal for both accuracy and communication speed.},
}

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

@article {pmid42911846,
year = {2026},
author = {Shen, Q and Eliseyev, A and Kansara, V and Neves Briard, J and Song, YL and Carmona, J and Gonzales, IJ and Velazquez, A and Klein, AJ and Agarwal, S and Park, S and Ghoshal, S and Roh, D and Connolly, ES and Claassen, J},
title = {Clinically unresponsive patients with acute brain injury activate a brain-computer interface.},
journal = {Brain : a journal of neurology},
volume = {},
number = {},
pages = {},
doi = {10.1093/brain/awag210},
pmid = {42911846},
issn = {1460-2156},
support = {NS112760/NS/NINDS NIH HHS/United States ; NS106014/NS/NINDS NIH HHS/United States ; //Paris Brain Institute America (PBIA)/ ; MP-200963/CAPMC/CIHR/Canada ; //Fondation du Centre hospitalier de l'Université de Montréal/ ; //Power Corporation of Canada Chair in Neuroscience/ ; //Bourse Perras, Perras et Cholette de la Faculté de médecine de l'Université de Montréal/ ; //Royal College of Physicians and Surgeons of Canada/ ; },
abstract = {Brain activation to verbal commands can be detected by EEG in clinically unresponsive patients. These patients with cognitive motor dissociation (CMD) are unable to communicate with the environment despite preserved consciousness. Advanced brain-computer interfaces (BCI) are promising to support motor function in patients with motor neuron disease or spinal cord injury. Our primary objective was to determine if clinically unresponsive patients can control a self-paced, one-way BCI early after acute brain injury. Secondarily, we assessed the association of BCI control with short- and long-term outcomes. We conducted a prospective single-centre cohort study in 41 clinically unresponsive, critically ill and acutely brain injured patients, as well as in a convenience sample of 8 healthy volunteers and 22 conscious critically ill patients. The self-paced BCI analysed EEG in real time and modulated frequency and pitch of a tone that patients controlled by attempting a motor task: patients were provided with instructions 'to make the beep sound faster, keep opening and closing your right hand' alternating with 'to make the beep sound slower, stop opening and closing your right hand'. CMD was detected by applying machine learning to EEG recorded while patients listened to motor commands. Clinical outcomes were recovery of consciousness by hospital discharge, assessed with the Coma Recovery Scale-Revised, and recovery of functional independence at 12 months, defined as a Glasgow Outcome Scale-Extended score of 4 or higher (denoting the ability to function independently for 8 h or more). Additionally, we invited conscious patients as well as family members of conscious and clinically unresponsive patients to complete a survey on perceived utility and challenges associated with the BCI. In this study, 13 clinically unresponsive patients (32%) achieved BCI control and 8 of 41 (20%) were CMD-positive. All healthy volunteers and nine conscious patients (41%) were able to control the BCI. Median signal delay for BCI control was 49.8 [interquartile range (IQR) 39.9-75.6] seconds for clinically unresponsive patients, 17.1 (IQR 7.8-19.9) for conscious patients and 15.5 (IQR 12.9-21.2) for healthy volunteers. In the group of 41 clinically unresponsive patients, 4 of 13 patients with (31%) and 1 of 28 without BCI control (4%) recovered clinical consciousness at hospital discharge [odds ratio (OR) 21.2; 95% confidence interval (CI) 2.3-930.7]. At 12 months, 6 of 13 patients with (46%) and 2 of 28 without BCI control (7%) recovered functional independence (OR 9.9; 95% CI 1.8-81.6). None of the patients that were unable to control the BCI and were CMD-negative recovered. Patients with BCI control had shorter times to recovery of functional independence by 12 months than those who did not (hazards ratio 6.0; 95% CI 1.2-30.6). All respondents of the BCI survey (n = 37) expressed strong support for the potential future use of BCI devices. These results suggest that BCI control is a strong predictor of outcomes following acute brain injury and provide initial evidence that a bedside, EEG-based BCI system might permit bidirectional communication in critically ill unresponsive patients.},
}

@article {pmid42912743,
year = {2026},
author = {Zhu, Y and Shi, Y and Wu, J and Wei, Q},
title = {Decoding motor imagery EEG signals by a deep learning network integrating multiscale filtering and temporal-channel attention mechanisms.},
journal = {PloS one},
volume = {21},
number = {10},
pages = {e0361026},
pmid = {42912743},
issn = {1932-6203},
mesh = {*Deep Learning ; Humans ; *Electroencephalography/methods ; *Brain-Computer Interfaces ; Convolutional Neural Networks ; Signal Processing, Computer-Assisted ; },
abstract = {Due to their low signal-to-noise ratio and non-stationarity, accurately classifying EEG signals is still one of the main hindrances that limit the real-world applications of a brain-computer interface (BCI). Deep learning (DL) is a potential method for addressing the problem. However, how to create a suitable DL network for motor imagery (MI)-based BCIs poses severe challenges in the case of insufficient training data. In this paper, a DL network based on multiscale convolutional neural network (CNN) and temporal and channel attention mechanisms is proposed for decoding MI tasks. Temporal and spatial features are extracted by four parallel CNN branches, each of which is followed by a Transformer block for learning the global correlations of local time features. The outputs from the four Transformer blocks are concatenated along channel dimension and then input into a channel attention module for channel fusion. The subsequent feature map is classified by a CNN, which replaces the commonly used fully connected layer for reducing computational complexity. The proposed network was evaluated on three publicly available BCI datasets and compared with five state-of-the-art DL networks. The training data sets are augmented using the approach segmentation and reconstruction. The experimental results demonstrate that the proposed network achieved an average accuracy of 80.21% on Dataset 1 (4-class tasks), 90.19% on Dataset 2 (2-class tasks), and 97.32% on Dataset 3 (4-class tasks), surpassing the strongest baselines FBCNet on Dataset 1, MSFBCNN on Dataset 2 and MSFBCNN on Dataset 3 by 2.59%, 1.27%, and 1.99% respectively. The network effectively decoded the MI-based EEG signals, providing a potential method for BCI applications.},
}

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

@article {pmid42913013,
year = {2026},
author = {Wu, Y and Chang, T and Chen, S and Quan, Y and Sun, Q and Van Hulle, MM and Liu, Y and Jiang, N and Farina, D and Yang, Y and Mao, Q and He, J},
title = {Biophysically constrained learning from healthy neural priors improves motor decoding in pathological cortical networks.},
journal = {Science advances},
volume = {12},
number = {43},
pages = {eadx5564},
pmid = {42913013},
issn = {2375-2548},
mesh = {Humans ; *Motor Cortex/physiopathology/pathology/physiology ; *Brain-Computer Interfaces ; Electrocorticography ; *Learning ; *Nerve Net/physiopathology ; *Glioma/physiopathology ; },
abstract = {High-performance brain-computer interfaces (BCIs) rely on stable neural representations, yet brain functional impairments inherently disrupt canonical encoding, creating a fundamental decoding paradox. Conventional data-driven approaches struggle to bridge this gap because of pathological data scarcity and heterogeneity. Here, we propose to leverage healthy neural priors and introduce Pathology-Aware Consensus Training (PACT), a paradigm embedding biophysical constraints into healthy neural recordings to guide neural decoding. In a gesture classification task using high-density electrocorticography from patients with gliomas infiltrating the motor cortex, PACT improved the accuracy by more than 10%, effectively restoring the distorted semantic geometry of motor commands and realigning them with healthy canonical representations (correlation increase: r = 0.66 to r = 0.94). Mechanistically, PACT induced an emergent "inverse weighting" strategy: The model autonomously suppresses high-power pathological noise while reanchoring spatial attention to functional compensatory regions. This study provides a proof of concept for improving decoding in compromised networks by harnessing healthy neural dynamics, laying the theoretical foundation for future BCI deployments in clinical populations.},
}

Humans
*Motor Cortex/physiopathology/pathology/physiology
*Brain-Computer Interfaces
Electrocorticography
*Learning
*Nerve Net/physiopathology
*Glioma/physiopathology

@article {pmid42913639,
year = {2026},
author = {Wang, Z and Tang, Q and Li, K and Li, J and Mou, J and Sun, X and Wei, Y and Chen, Y and Guo, P and Jin, L and Liu, B and Xiao, B and Duan, S and Dai, W and Ni, JD},
title = {A vagal gut-brain axis for feeding-induced gastric secretion in mice.},
journal = {Neuron},
volume = {},
number = {},
pages = {},
doi = {10.1016/j.neuron.2026.09.033},
pmid = {42913639},
issn = {1097-4199},
abstract = {Feeding begins with food-seeking and ends with nutrient absorption and waste elimination. Although meal-related sensory cues shape feeding behavior, how they engage neural circuits to control digestion remains unclear. We dissected vagal gut-brain-gut pathways mediating meal-induced gastric secretion. Stomach-innervating vagal afferents were necessary and sufficient for secretion. Among them, Glp1r[+] and Sst[+] neurons drove acid, pepsin, and gastrin secretion, whereas Oxtr[+] and Vip[+] neurons had little effect. Piezo2 was expressed in Glp1r[+] mechanosensory afferents, and conditional deletion of Piezo2 attenuated distension-evoked secretion. On the efferent limb, genetically defined vagal motor-neuron subsets showed distinct secretory effects: Cck[+] neurons broadly increased acid, pepsin, and gastrin, whereas Grp[+] neurons preferentially affected acid secretion. Finally, enteric circuit dissection revealed modular output control, with Calb2[+] neurons driving secretion with minimal motility effects. Together, these results define a cell-type-resolved vagal-enteric architecture for gastric control.},
}

@article {pmid42913685,
year = {2026},
author = {Washizawa, Y},
title = {Training Free Signal Enhancement by Random Sampling for Steady State Visual Evoked Potential BCI.},
journal = {Biomedical physics & engineering express},
volume = {},
number = {},
pages = {},
doi = {10.1088/2057-1976/aeb80c},
pmid = {42913685},
issn = {2057-1976},
abstract = {The steady-state visual evoked potential (SSVEP) is widely used for electroencephalography (EEG) based brain computer interfaces (BCIs). In order to determine the target commands for SSVEP-BCIs, deep neural networks have been widely used. However, learning-based approaches require training data, and the upfront costs are high. This study proposes a training free signal enhancement method for SSVEP-BCIs. The proposed method enhances SSVEP components by repeatedly performing random sampling and pasting of EEG signals. Our experiments using synthesized signals and open benchmark datasets demonstrate the performance of the proposed method. The proposed method outperforms conventional methods for all signal lengths, demonstrating an improvement in information transfer rate of up to 11.30 bit/min. .},
}

@article {pmid42900612,
year = {2026},
author = {Bhagat, V and Yadav, RK and Ghosh, S and Muthukrishnan, SP and Bo, APL and Roy, S and Dobbins, C},
title = {EEG-Based Characterization of Cortical Dynamics during Graded Resistance Exercise.},
journal = {IEEE journal of biomedical and health informatics},
volume = {PP},
number = {},
pages = {},
doi = {10.1109/JBHI.2026.3745288},
pmid = {42900612},
issn = {2168-2208},
abstract = {Understanding how brain activity varies during voluntary movement under different resistance conditions may provide insights into neural responses associated with physical demands and inform the development of neurorehabilitation strategies. In this study, we investigated changes in neural activity associated with increasing resistance levels during unilateral bicep curl movement. Ten healthy participants performed bicep curls under five resistance conditions while their cortical activity was recorded using electroencephalography (EEG). Movement-related cortical potentials (MRCPs), power spectral density (PSD) topographies, nonparametric spatiotemporal cluster-based permutation tests and linear mixed-effects models (LMMs) were used to characterize EEG differences across resistance conditions. Differences in MR-CPs and frequency-specific power spectral density (PSD) patterns were observed across resistance conditions, with significant spatiotemporal clusters identified during pre-movement, peak contraction, and movement execution phases. A Triplet Temporal Convolutional Network (TTCN) was also employed to learn latent representations from these EEG signals, and its discriminative capability was evaluated using a k-nearest neighbors (KNN) classifier. The learned embeddings showed separation between the resistance conditions, with classification errors occurring primarily between neighboring conditions. Using a within subject leave-one-trial-out (LOTO) evaluation strategy, the TTCN-KNN framework achieved accuracy of 97.2%. Furthermore, subject-independent evaluation were conducted using a leave-one-subject-out (LOSO) approach, achieving a maximum accuracy of 93.0%. These findings suggest that EEG signals contain discriminative information for distinguishing the resistance conditions in both within-subject and subject-independent settings, providing preliminary in sights for future EEG-based monitoring of motor effort and adaptive brain-machine interface applications.},
}

@article {pmid42903277,
year = {2026},
author = {Kaur, B and Middha, S and Ali, A and Khan, N and Alhejaili, R and Almutairi, AH},
title = {An attention-enhanced hybrid CNN-BiLSTM framework for multiclass epileptic seizure recognition using EEG signals.},
journal = {Frontiers in computational neuroscience},
volume = {20},
number = {},
pages = {1949040},
pmid = {42903277},
issn = {1662-5188},
abstract = {INTRODUCTION: Electroencephalogram (EEG) signal classification plays a pivotal role in brain-computer interfaces (BCIs) and the diagnosis of neurological disorders. However, the non-stationary, non-linear, and low signal-to-noise ratio characteristics of EEG signals make accurate feature extraction challenging. Conventional deep learning approaches often struggle to simultaneously capture localized transient patterns, deep temporal dependencies, and long-range contextual relationships within EEG signals.

METHODS: To address these challenges, an optimized end-to-end hybrid deep learning framework, termed CNN-BiLSTM-MHSA, is proposed for multiclass epileptic seizure recognition. The architecture sequentially integrates a multilayer one-dimensional Convolutional Neural Network (CNN) for localized feature extraction, a Bidirectional Long Short-Term Memory (BiLSTM) network for capturing forward and backward temporal dependencies, and a Multi-Head Self-Attention (MHSA) mechanism for dynamically emphasizing discriminative neurological events. Robust preprocessing includes Butterworth bandpass filtering, Z-score normalization, and data augmentation using Gaussian noise injection and amplitude scaling. The framework was evaluated on the UCI/Kaggle Epileptic Seizure Recognition dataset comprising 11,500 EEG segments distributed across five classes.

RESULTS: The proposed framework achieved an overall accuracy of 98.83%, precision of 98.95%, recall (sensitivity) of 98.84%, F1-score of 98.90%, specificity of 99.71%, balanced accuracy of 98.84%, ROC-AUC of 99.80%, and PR-AUC of 99.72%. Class-wise accuracies exceeded 99% for most categories, with one seizure class achieving perfect recognition. Comparative evaluation further demonstrated that the proposed framework outperformed several recent state-of-the-art EEG seizure classification methods.

DISCUSSION: The superior performance can be attributed to the complementary representation capabilities of the hybrid architecture. CNN layers effectively learn localized signal patterns, BiLSTM captures bidirectional temporal dependencies, and MHSA enhances the representation of long-range contextual relationships by assigning greater importance to informative temporal regions. The combination of robust preprocessing and data augmentation further contributes to improved generalization and classification reliability.},
}

@article {pmid42904129,
year = {2026},
author = {Li, X and Ni, H and Cui, G and Wei, C and Zhu, Y},
title = {Impact of motor guided imagery therapy on stroke patients with pain symptoms: a three-level meta-analysis and systematic review of randomized controlled trials.},
journal = {Frontiers in neurology},
volume = {17},
number = {},
pages = {1920523},
pmid = {42904129},
issn = {1664-2295},
mesh = {Humans ; *Imagery, Psychotherapy/methods ; *Stroke/complications ; Randomized Controlled Trials as Topic ; *Stroke Rehabilitation/methods ; *Pain Management/methods ; *Pain/etiology ; },
abstract = {BACKGROUND: Post-stroke pain is a common and intractable complication that hinders neurorehabilitation progress. Motor Guided Imagery (MGI), an emerging rehabilitation technique based on brain-computer interaction principles, has shown potential in relieving post-stroke pain, yet high-quality evidence on its multi-dimensional efficacy remains insufficient. This study aimed to systematically evaluate the effects of MGI on pain, sensorimotor function, sleep quality, life status, and psychological mood in stroke patients using a three-level meta-analysis.

OBJECTIVE: This study aimed to quantitatively assess the therapeutic effects of Motor Guided Imagery for stroke-related pain and explore potential intervention moderators using a three-level meta-analysis framework.

METHODS: Eight Chinese and English databases were searched from inception to June 29, 2025, to collect randomized controlled trials (RCTs) of MGI for stroke patients with pain. A three-level random-effects model was adopted to pool Hedges' g effect sizes, accompanied by heterogeneity testing, moderator analysis, influence analysis, and publication bias assessment. Influence-based outlier detection was implemented; analyses retaining all effect-size observations were treated as the primary main analysis, while outlier-removed analyses were reported only as supplementary sensitivity checks.

RESULTS: Motor-guided imagery training significantly relieved pain (g = -0.524, 95% CI -0.779 to -0.269, p = 0.0006; negative Hedges' g indicates clinical improvement for this outcome) and improved sleep quality (g = -1.658, 95% CI -2.115 to -1.201, p < 0.001; negative Hedges' g indicates clinical improvement for this outcome) in stroke patients. Meanwhile, it significantly enhanced motor function (g = 1.395, 95% CI 0.963 to 1.826, p < 0.001), life status (g = 1.757, 95% CI 0.966 to 2.549, p = 0.0012), and psychological-emotional status (g = 1.729, 95% CI 1.045 to 2.413, p = 0.0002); positive Hedges' g indicates clinical improvement for the latter three outcomes. Moderator analysis indicated that acute-phase intervention (≤4 weeks) and high-frequency regimens (>5 sessions/week) tended to produce better analgesic effects, while age, gender, single-session duration, and pain location had no significant moderating effects.

CONCLUSION: MGI may produce beneficial rehabilitation effects that can alleviate post-stroke pain and synergistically improve motor function, sleep, daily-living ability, and psychological state of stroke patients. Exploratory subgroup trends indicated potentially larger effects for early implementation within the acute phase and high-frequency short-duration regimens. However, these observations should be interpreted as hypothesis-generating clues given non-significant overall moderator tests. Further high-quality research is required to confirm these findings.

REGISTRATION: The systematic review protocol was retrospectively registered with PROSPERO under registration ID CRD420261278812, after literature search and full-text screening had been completed.

https://www.crd.york.ac.uk/PROSPERO/view/CRD420261278812; identifier: CRD420261278812.},
}

Humans
*Imagery, Psychotherapy/methods
*Stroke/complications
Randomized Controlled Trials as Topic
*Stroke Rehabilitation/methods
*Pain Management/methods
*Pain/etiology

@article {pmid42904507,
year = {2026},
author = {Zhang, K and Lin, J and Tong, P and Liu, B and Li, J and Tian, H and Chen, J and Hong, Z and Wu, Y and Huang, Y},
title = {Realizing Ultrahigh Polarization and Excellent Thermal Stability in Piezoceramics via Defect-Stress Engineering.},
journal = {Small (Weinheim an der Bergstrasse, Germany)},
volume = {},
number = {},
pages = {e76395},
doi = {10.1002/smll.76395},
pmid = {42904507},
issn = {1613-6829},
support = {52572143//National Natural Science Foundation of China/ ; U25A20231//National Natural Science Foundation of China/ ; 92463306//National Natural Science Foundation of China/ ; LD24E020003 LR25E020003//Zhejiang Provincial Natural Science Foundation of China/ ; 2023R5231//Special Support Plan for High Level Talents in Zhejiang Province/ ; //Open Research Project of Innovation Center of Yangtze River Delta of Zhejiang University/ ; 2024ZY01019//Science and Technology Department of Zhejiang Province/ ; },
abstract = {Large polarization is critical for improving the piezoelectric and actuation performance of ferroelectric materials. As a promising lead-free candidate, BiFeO3-BaTiO3 inherits the excellent polarization capability of BiFeO3. However, the actual polarization achieved in its ceramics remains far below the theoretical predictions. In this work, we propose a novel defect-stress engineering strategy to enhance the polarization substantially. By intentionally introducing A-site vacancies into 0.70BiFeO3-0.30BaTiO3 ceramics, combined with quenching treatment, localized stress-concentration regions are formed within the ceramic matrix. This strategy successfully achieves a remarkable increase in polarization displacement and promotes the formation of polar nanoregions. Under an electric field of 60 kV/cm, the optimized ceramics exhibit a high remanent polarization of 46.40 µC/cm[2], which is significantly superior to that of most reported lead-free counterparts. Moreover, the unipolar strain shows only a 13% variation from room temperature to 180°C, demonstrating excellent thermal stability. This work confirms that defect-stress engineering is an effective approach to enhance polarization in bulk lead-free ferroelectrics and provides a new framework for developing lead-free ferroelectric materials with both high polarization and superior temperature stability.},
}

@article {pmid42905181,
year = {2026},
author = {Bao, C and Hu, R},
title = {Integrated transcriptomic analysis identifies an AKT1-associated metabolic-inflammatory signature in osteoporosis.},
journal = {Clinical and experimental medicine},
volume = {26},
number = {1},
pages = {},
pmid = {42905181},
issn = {1591-9528},
mesh = {*Osteoporosis/genetics/metabolism ; Humans ; *Proto-Oncogene Proteins c-akt/genetics/metabolism ; Gene Expression Profiling ; Animals ; Fatty Acids/metabolism ; Machine Learning ; *Inflammation/genetics ; *Transcriptome ; Female ; Mice ; Bone Density ; },
abstract = {Osteoporosis is a significant global health challenge, and low bone mineral density (Low-BMD) is a major clinical feature and risk indicator of osteoporosis. Fatty acid metabolism has been increasingly implicated in bone remodeling and inflammatory regulation, but the transcriptomic links between fatty acid metabolism and immune-inflammatory alterations in Low-BMD remain unclear. We integrated three GEO datasets, including 60 High-BMD controls and 59 Low-BMD subjects. Differential expression analysis, immune infiltration analysis, NMF clustering, WGCNA, pseudotime analysis, and machine learning were performed to identify fatty acid metabolism-related genes associated with Low-BMD. AKT1 knockdown was used to evaluate its potential role in regulating inflammatory mediators. Eighteen fatty acid metabolism-related differentially expressed genes were identified between High-BMD and Low-BMD samples and were enriched in inflammatory, immune, and osteoclast-related pathways. NMF clustering revealed two Low-BMD molecular subtypes with distinct metabolic and immune-inflammatory features. Machine learning identified an AKT1-containing diagnostic signature with good predictive performance, and AKT1 was positively correlated with inflammatory response scores and inflammatory markers. Quantitative Real-time PCR confirmed Akt1 knockdown, and ELISA showed reduced IL-6, TNF-α, and IL-1β secretion in MC3T3-E1 cells. AKT1 may serve as a fatty acid metabolism-related candidate gene associated with immune-inflammatory remodeling and inflammatory mediator regulation in Low-BMD. These findings suggest a potential AKT1-associated metabolic-inflammatory regulatory relationship and provide candidate biomarkers for future validation in osteoporosis-related bone loss.},
}

*Osteoporosis/genetics/metabolism
Humans
*Proto-Oncogene Proteins c-akt/genetics/metabolism
Gene Expression Profiling
Animals
Fatty Acids/metabolism
Machine Learning
*Inflammation/genetics
*Transcriptome
Female
Mice
Bone Density

@article {pmid42905589,
year = {2026},
author = {Mandal, R and Prabhakaran, GS and Kowshik, AV and Balaji, BM and Jayaprakash, B},
title = {A Sensitive BLE-Enabled Wearable fNIRS Biosensor for Real-Time Monitoring and Mapping of Task-Evoked Oxygen Influxes During On-Screen Cognitive Tasks.},
journal = {IEEE transactions on bio-medical engineering},
volume = {PP},
number = {},
pages = {},
doi = {10.1109/TBME.2026.3745751},
pmid = {42905589},
issn = {1558-2531},
abstract = {OBJECTIVE: To develop and validate a compact Bluetooth Low Energy (BLE)-enabled wearable functional near-infrared spectroscopy (fNIRS) device for real-time hemodynamic monitoring and proposes a task-evoked metric for identifying transient oxygenation events during cognitive engagement.

METHOD: We designed a four-channel wearable fNIRS biosensor that integrates optical sensing, signal acquisition, and wireless transmission into a compact, cable-free platform using off-the-shelf electronic components. The device was characterized through optical power linearity, ADC voltage input, pulsed light acquisition, and signal-to-noise ratio analysis. Its performance was validated using forearm vascular occlusion and breath-hold experiments, followed by cortical recordings during breath-hold tasks, finger tapping, dynamic visual stimulation, and a screen-based spatial learning and memory task. We further formulated task-evoked instances of differential oxygen influx (tEIDO), computed from the instantaneous differential response between $\Delta$[HbO] and $\Delta$[Hb], as a hemodynamic index of task-locked neurovascular activity.

RESULT: The device reliably detected expected oxy- and deoxy-hemoglobin changes during vascular and breath-hold validation experiments and captured task-evoked responses over motor, visual, and prefrontal cortices. The tEIDO metric identified transient differential oxygenation events during visual stimulation and revealed task-aligned oxygenation dynamics. During spatial navigation, differential oxygenation maps showed hemodynamic modulation around the target region only after memory acquisition.

SIGNIFICANCE: The proposed BLE-enabled fNIRS platform, together with tEIDO, offers a compact and sensitive approach for untethered real-time cortical hemodynamic monitoring, with potential applications in mobile neuroimaging, neuroergonomics, cognitive-state monitoring, and brain-computer interface systems.},
}

@article {pmid42906532,
year = {2026},
author = {Nagarajan, SS and Khanna, P},
title = {Optimizing learning for control of human-machine interfaces.},
journal = {Neuron},
volume = {114},
number = {20},
pages = {3734-3735},
doi = {10.1016/j.neuron.2026.09.023},
pmid = {42906532},
issn = {1097-4199},
mesh = {Humans ; *Brain-Computer Interfaces ; *Learning/physiology ; *Feedback, Sensory/physiology ; *Man-Machine Systems ; *User-Computer Interface ; },
abstract = {Human-machine interfaces, such as augmented reality devices, hand prostheses, and brain-computer interfaces, are increasingly being used in neurorehabilitation. In this issue of Neuron, Vassiliadis et al.[1] examine the interaction between closed-loop continuous reinforcement and sensory feedback in learning of human-machine interface control.},
}

Humans
*Brain-Computer Interfaces
*Learning/physiology
*Feedback, Sensory/physiology
*Man-Machine Systems
*User-Computer Interface

@article {pmid42906547,
year = {2026},
author = {Zhang, YY and Hu, MT and Xie, SZ and Yu, CX and Xu, YH and Chen, X and Tang, MY and Li, CY and Zhu, Y and Hu, JX and Tan, LH and Wang, H and Yang, H and Yu, YQ and Pan, BX and Li, XM and Lian, H},
title = {A GABAergic preoptic-raphe circuit mediates paradoxical coupling of anxiety and social preference after comparative social deficit.},
journal = {Neuron},
volume = {},
number = {},
pages = {},
doi = {10.1016/j.neuron.2026.09.032},
pmid = {42906547},
issn = {1097-4199},
abstract = {Social disconnection is a major risk factor for anxiety disorders, the most prevalent mental illnesses worldwide, yet its neural substrates remain unclear. Here, we establish a comparative social deficit (CSD) model in adult male mice that induces a paradoxical phenotype of increased anxiety-like behavior together with heightened social preference, a pattern conceptually similar to human loneliness. Mechanistically, CSD increases the excitability of medial preoptic GABAergic (mPOA[Gad2]) neurons, which bidirectionally regulate CSD-induced behavioral changes. Transcriptomic and electrophysiological analyses identify downregulation of the Kv4.3 potassium channel (Kcnd3) as a molecular driver of mPOA[Gad2] hyperexcitability. Furthermore, mPOA[Gad2] neurons exert monosynaptic inhibition onto dorsal raphe serotonergic (DRN[5-HT]) neurons. This inhibitory projection, strengthened by CSD, mediates CSD-induced behavioral effects. These findings define a molecular and circuit mechanism by which social disconnection remodels mPOA-DRN interactions to drive the paradoxical coupling of anxiety and elevated social preference, providing a translational framework for understanding loneliness-related psychiatric vulnerability.},
}

@article {pmid42906584,
year = {2026},
author = {Yang, T},
title = {ACF-Net: Design of an adaptive cross-modal fusion network for cross-subject wearable EEG-EMG gesture recognition.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/aeb723},
pmid = {42906584},
issn = {1741-2552},
abstract = {Cross-subject gesture recognition in wearable brain-computer interfaces is constrained by four structural deficiencies: symmetric cross-modal attention discarding cortical precedence over muscular execution, priors confined to post hoc analysis, invariance anchored on unimodal features, and adaptation requiring labelled target-domain data. This study closes these gaps by embedding corticomuscular coupling directly into the network architecture rather than recovering it from data. Approach: ACF-Net is a four-stage pipeline embedding neurophysiological priors directly into fusion. The corticomuscular coherence prior injection layer (CMC-Prior) encodes time-lagged coherence and phase-locking values for every electroencephalography (EEG)-electromyography (EMG) channel pair as a learnable bias tensor. Granger-causality-inspired directed cross-modal attention (GC-Attn) uses a causal mask allowing only EEG-to-EMG influence within 30-120 ms. The fused embedding is projected onto a symmetric positive definite manifold with log-Euclidean triplet loss. Physiology-consistent test-time adaptation (PC-TTA) updates only batch-normalisation affine parameters, using EEG-precedes-EMG as unlabelled self-supervision, eliminating labelled target data. Leave-one-subject-out validation on WAY-EEG-GAL (12 participants) and WEE-Ges (20 participants, three days) used fold-wise priors and Holm-corrected Wilcoxon tests. Main results: ACF-Net reached 87.6% on WAY-EEG-GAL and 89.2% on WEE-Ges, exceeding the strongest baseline by 9.4 and 9.1 percentage points with all comparisons significant after correction. Ablation attributed gains of 6.3, 4.9, 3.4 and 1.8 percentage points to directed attention, prior injection, manifold embedding and test-time adaptation. Accuracy remained 77.5% under 20 mm simulated electrode displacement, adaptation reduced the cross-day loss from 6.8 to 2.1 percentage points, and a forward pass on an NVIDIA Jetson Nano took 32.4 ms within a 178 ms end-to-end response at 30.9 Hz. Significance: Learned coupling weights recovered central-to-forearm-extensor projection, and delays matched corticospinal conduction with motor-unit recruitment, suggesting physiological structure, not dataset-specific statistics. Yet healthy young adults, gel-based/semi-dry wired hardware, and simulated displacement make this methodological; clinical and wireless dry-electrode transfer remains unestablished.},
}

@article {pmid42906586,
year = {2026},
author = {Podmore, JJ and Radman, M and Poli, R and Paulmann, S and Daly, I},
title = {Towards semantic brain-computer interfacing: The graded inventory of semantic triggers (GIST) paradigm.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/aeb724},
pmid = {42906586},
issn = {1741-2552},
abstract = {Semantic Brain-Computer Interfaces (sBCIs) aim to decode conceptual information directly from neural activity, enabling more efficient and natural communication for individuals with severe motor impairments. However, existing approaches are often confounded by stimulus-driven perceptual processes and lack a reliable means of tagging the moment of concept formation. We introduce the Graded Inventory of Semantic Triggers (GIST), a novel electroencephalography (EEG) paradigm designed to elicit semantic retrieval from graded verbal clues with precise temporal alignment. Approach: EEG data were recorded from 14 participants performing the GIST task across three semantic categories (food, musical instruments, and body parts). Target neural activity was aligned to each participant's self-reported moment of semantic retrieval (spacebar press), enabling response-locked analysis. Here, we propose a staged framework where the first comprises a Conformer-based encoder trained on pooled data using a hybrid contrastive objective that combines instance-level consistency and cross-participant class alignment with domain-adversarial learning to reduce participant-specific information. In the second stage, the frozen encoder is transferred to a held-out participant whose data played no part in Stage 1, and a support vector machine is calibrated on that participant's own labelled items under a nested leave-one-item-out protocol. Results: Above-chance semantic decoding was achieved in all three pairwise tasks. Mean accuracies were 73.4% (Body vs. Food), 70.7% (Body vs. Music), and 68.2% (Food vs. Music). Participant-level permutation tests were significant in 9/11 participants for Body vs. Food and 8/11 for each of the other tasks (p < 0.05). The pre-response interval at [-1.5s, -0.5s] was the most discriminative. Significance: The GIST paradigm enables controlled elicitation and temporal alignment of semantic processing while reducing perceptual confounds. Combined with cross-participant representation pretraining and participant-specific calibration, the results demonstrate above-chance decoding of semantic categories from EEG in individuals excluded from representation learning, supporting the feasibility of sBCIs.},
}

@article {pmid42907939,
year = {2026},
author = {Zheng, ZW and Wu, ZY},
title = {Preclinical models of Wilson's disease: From pathogenesis to clinical translation.},
journal = {Zoological research},
volume = {47},
number = {6},
pages = {1853-1864},
doi = {10.24272/j.issn.2095-8137.2025.679},
pmid = {42907939},
issn = {2095-8137},
mesh = {*Hepatolenticular Degeneration/pathology/genetics/therapy ; Animals ; Humans ; *Disease Models, Animal ; Copper-Transporting ATPases/genetics ; Translational Research, Biomedical ; },
abstract = {Wilson's disease (WD) is an autosomal recessive disorder of copper homeostasis caused by pathogenic ATP7B variants, leading predominantly to hepatic and neurological injury. Extensive heterogeneity in genotype, age at onset, clinical presentation, and treatment response complicates both diagnosis and therapeutic development, highlighting the need for reliable experimental models that reproduce distinct features of WD pathogenesis. This review critically evaluates animal models, conventional cell systems, induced pluripotent stem cell models, and organoids of WD, with an emphasis on the phenotypes reproduced, experimental strengths and limitations, and translational applications of each platform. Rodent models have been instrumental in defining mechanisms of hepatic copper accumulation and toxicity and remain widely used for preclinical therapeutic evaluation, but existing models do not adequately reproduce the neurological manifestations characteristic of human WD. Patient-derived cells and organoids provide complementary platforms for investigating mutation-specific pathobiology, individualized drug responses, and genome-editing strategies, although incomplete maturation, limited cellular complexity, and insufficient standardization remain important constraints. Future progress will depend on models that more faithfully capture neurological disease, greater use of humanized systems, improved physiological fidelity of organoids, and standardized criteria for phenotypic assessment. Integrating complementary models across the preclinical pipeline should strengthen mechanistic insight and accelerate development of precision therapies for WD.},
}

*Hepatolenticular Degeneration/pathology/genetics/therapy
Animals
Humans
*Disease Models, Animal
Copper-Transporting ATPases/genetics
Translational Research, Biomedical

@article {pmid42908451,
year = {2026},
author = {Yin, L and Ye, Y and Li, Y and Luo, J and Zhou, H and Tian, Y and Luo, R},
title = {[Discussion on the closed-loop regulation strategy for acupuncture "Deqi" based on brain-computer interface and its peripheral-central mechanisms].},
journal = {Zhen ci yan jiu = Acupuncture research},
volume = {51},
number = {10},
pages = {1354-1366},
doi = {10.13702/j.1000-0607.20260385},
pmid = {42908451},
issn = {1000-0607},
mesh = {Humans ; *Brain-Computer Interfaces ; *Acupuncture Therapy ; Electroencephalography ; *Brain/physiology ; },
abstract = {It is held that "Deqi" is the key to the therapeutic effect of acupuncture, but its evaluation has long relied on subjective feelings and lacks objective biological indicators. In the present article, we systematically review the current research status of brain-computer interface (BCI) technology based on electroencephalogram (EEG) in analyzing the peripheral-cerebral neural mechanism of "Deqi", and further explore how to utilize the closed-loop regulation ability of BCI to actively induce and maintain the "Deqi" state. By integrating domestic and foreign literature, we analyzed the peripheral initiation mechanism and central response characteristics of "Deqi", and explored its application value of EEG-BCI in the objective identification and the construction of closed-loop feedback systems. Studies have demonstrated that EEG, with its millisecond-level time resolution, can capture the dynamic changes of neural oscillations during the "Deqi" process in real time. The feature decoding method based on machine learning further realizes the objective identification of the "Deqi" state, while the closed-loop feedback system constructed by BCI technology provides an innovative path for precise acupuncture treatment. However, this field still faces some challenges such as signal artifact interference, insufficient decoding accuracy, and inconsistent research paradigms. The deep integration of EEG-BCI with acupuncture medicine provides a key technical platform for decoding the scientific connotation of "Deqi" and active regulation. In the future, efforts should be made to develop dynamic network decoding technology, further deepen the research on peripheral-cerebral coupling mechanisms, and promote the transformation of the system towards intelligence and standardization.},
}

Humans
*Brain-Computer Interfaces
*Acupuncture Therapy
Electroencephalography
*Brain/physiology

@article {pmid42908918,
year = {2026},
author = {Kodama, T and Kitai, K and Mizomoto, Y and Nishigaya, K and Hashimoto, S and Hiraga, R and Shimizu, S and Kawabata, M and Hirabayashi, Y and Ikesue, K and Kako, H and Yasui, S and Ito, H and Nishimoto, K and Katayama, O and Yamaguchi, H and Nagano, H and Inoue, M and Murata, S and Morita, K},
title = {Graded multisensory neurofeedback via the iNems system for comprehensive upper limb recovery after stroke: a multiple baseline study.},
journal = {Frontiers in human neuroscience},
volume = {20},
number = {},
pages = {1938622},
pmid = {42908918},
issn = {1662-5161},
abstract = {INTRODUCTION: Post-stroke upper-limb recovery remains challenging, especially in the presence of sensory impairments. Conventional brain-computer interfaces (BCIs) are constrained by binary feedback, single-modality stimulation, and exclusion of patients with sensory deficits. We developed the Imagery Neurofeedback-based Multi-sensory System (iNems), a neurofeedback system that provides graded multisensory (visual and electrical) feedback based on ERD magnitude. This study validated iNems' efficacy in motor function, self-reported numbness, and cognitive functions using a multiple-baseline design (MBD).

METHODS: Four participants with post-stroke upper-limb motor and sensory impairments underwent a multiple baseline design. The baseline phase (phase A) used random feedback; the intervention (phase B) used adaptive iNems. iNems quantified motor intention via sensorimotor EEG event-related desynchronization (ERD) and delivered graded visual and electrical feedback. Primary outcomes were the Fugl-Meyer Assessment-Upper Extremity (FMA-UE) and Box and Block Test (BBT). Secondary outcomes included body cognition, numbness, and neurophysiological (ERD/ERS) changes.

RESULTS: All participants demonstrated significant improvement during the intervention. FMA-UE increased by 6-16 points. The BBT showed large effects (integrated Tau-U = 0.75). Body cognition improved significantly in both sense of agency (Tau-U = 0.80) and ownership (Tau-U = 0.89). Numbness was reduced in affected cases (Tau-U = -0.85). Neurophysiological changes correlated with functional gains, with some patterns suggesting increased neural efficiency.

CONCLUSION: This first MBD-validated demonstration shows that graded, multisensory graded neurofeedback based on ERD magnitude can improve motor, self-reported numbness, and cognitive domains after stroke. These findings support adaptive closed-loop therapy as a promising, multifaceted rehabilitation strategy.},
}

@article {pmid42909857,
year = {2026},
author = {Liu, C and Mao, Y and Ren, Z and Shang, Y and Xu, H and Feng, W and Zhou, J and Zhang, D and Cui, X and Hu, X},
title = {Conductive hydrogel bioelectronics for chronic neuromodulation: From biomimetic design and bioactive interfaces to integrated neural platforms.},
journal = {Materials today. Bio},
volume = {41},
number = {},
pages = {103722},
pmid = {42909857},
issn = {2590-0064},
abstract = {In biomedical engineering, hydrogel bioelectronics brings together the interface between flexible electronic devices and biological systems, opening up a new way for chronic neural regulation. Here we focus on the recent progress in this field, from material design to system integration. The design of high-performance conductive hydrogels is an entry point. The multi-level structure engineering strategy, such as the introduction of carbon nanotubes or conductive polymers, can raise the conductivity and mechanical toughness together. When biomimetic adhesion molecules such as polydopamine or tannic acid are added, the hydrogel has the ability of wet tissue adhesion, and anti swelling and active anti-inflammatory functions are also embedded in the material system. As a result, the resulting neural interfaces are not only more durable in practical applications but also ensure excellent biocompatibility. Turning to applications. How does this kind of material promote the development of chronic neuromodulation technology? Peripheral nerve repair, brain computer interface, spinal cord injury, autonomic nervous system regulation can be seen in several scenes. The long-term and high-precision recording and regulation of neural signals is exactly the most important link in these applications. Hydrogel electronics does have its own uniqueness in this regard. However, there is still a long way to go between laboratory research and clinical transformation. How to ensure long-term stability? Can controllable degradation be achieved in practice? Will the signal fidelity drop with time? How to optimize wireless operation and system integration? Large scale manufacturing is an unavoidable reality test. These directions are probably the worthy areas for follow-up research. Looking forward, if the lines of intelligent response materials, closed-loop sensing and control, and advanced micro processing technology can be screwed together, hydrogel bioelectronics may truly move towards the next generation of personalized, dynamic and adaptive neural diagnosis and treatment platform.},
}

@article {pmid42890887,
year = {2026},
author = {Tejesh, BSS and Bhaskar, K and Sivaji, M and Dakupati, R and Ram, SK and Sanki, PK and Ramakrishnan, M},
title = {A sliding window and Fourier decomposition method based 1D-CNN framework for motor imagery EEG classification.},
journal = {Frontiers in computational neuroscience},
volume = {20},
number = {},
pages = {1945160},
pmid = {42890887},
issn = {1662-5188},
abstract = {Electroencephalography (EEG) signal classification is fundamental requirement for effective motor imagery (MI) brain-computer interface (BCI) applications. However, current MI classification methods often depend on handcrafted spatial filtering or complex time-frequency transformations that increase computational complexity. To mitigate this limitation, we propose a unified architecture that integrates a sliding window (SW), a Fourier decomposition method (FDM), and a one-dimensional convolutional neural network (1D-CNN) for MI-EEG classification. The raw EEG data were segmented using the SW approach to augment the data and extract more temporal feature patterns from each trial. The segmented signal, containing time-domain features, was subsequently processed using a band-pass filter and FDM to extract frequency-based features. Here, FDM decomposes the signal into eight orthogonal Fourier Intrinsic Band Functions (FIBFs). A 1D-CNN architecture was employed to learn representations from the decomposed EEG signals. The proposed framework was evaluated on the BCI Competition III Dataset IVa (BCIC III IVa) and BCI Competition IV-2a (BCIC IV 2a) datasets. Our method achieved accuracies of 79.14% and 93.67% on the BCIC IV 2a and BCIC III IVa datasets, respectively, under the subject-dependent evaluation protocol. To evaluate cross-subject generalization, the proposed framework was validated using the Leave-One-Subject-Out (LOSO) protocol, achieving average classification accuracies of 76.49% on the BCIC III Dataset IVa and 55.50% on the BCIC IV 2a dataset. Furthermore, deployment on the NVIDIA Jetson Nano demonstrated complete pipeline latencies of 58.119 ms/trial and 85.055 ms/trial, corresponding to throughputs of 17.21 and 11.76 trials/s for the BCIC III IVa and BCIC IV 2a datasets, respectively. The results show the strong potential of the developed MI-EEG classification method to assist subjects with motor ailments.},
}

@article {pmid42890958,
year = {2026},
author = {Snowin, B and Hazeena, P and Narasimhan, L},
title = {Electroencephalography Changes in Emotional Recognition: A Signal Processing Approach Using Wavelet Features.},
journal = {Annals of neurosciences},
volume = {},
number = {},
pages = {09727531261483432},
pmid = {42890958},
issn = {0972-7531},
abstract = {BACKGROUND: motion recognition using electroencephalography (EEG) has gained significant attention in the fields of affective computing and human-computer interaction due to its ability to capture real-time neural responses. EEG signal variations associated with emotional states can be analysed using wavelet-based signal processing within the arousal-valence framework.

PURPOSE: This study aims to analyse EEG signal variations associated with emotional states using a wavelet-based signal processing approach within the arousal-valence framework.

METHODS: A prospective observational study was conducted on 48 healthy participants. Emotional states were induced using image, video and music stimuli categorised under the arousal-valence model. EEG signals were recorded using the standard 10-20 electrode placement system. The acquired signals were pre-processed and analysed using discrete wavelet transform (DWT) to extract time-frequency features, including energy and entropy. Statistical analysis was performed to evaluate hemispheric differences and gender-based variations.

RESULTS: Differences in EEG activity were observed between the left and right hemispheres across multiple emotional states (p < .05). Video and music stimuli produced stronger and more distinguishable EEG responses compared to image-based stimuli. Negative emotions demonstrated higher discriminative patterns than positive emotions. Gender-based variations were also observed in EEG responses.

CONCLUSION: Wavelet-based EEG feature analysis provides meaningful and discriminative information for emotion recognition and may provide a robust feature set for future machine learning-based emotion recognition systems. Further studies incorporating classification algorithms and larger data sets are required to validate their predictive performance.},
}

@article {pmid42893344,
year = {2026},
author = {Liu, J and He, Y and Liaw, HP and Russo, P and Gourdouparis, M and Huang, X and Chen, J and Liu, YH},
title = {A Compressive Neural Telemetry with Address-Event-μPacket Encoding and Duty-Cycling IR-UWB Transmitter.},
journal = {IEEE transactions on microwave theory and techniques},
volume = {74},
number = {8},
pages = {7429-7440},
pmid = {42893344},
issn = {0018-9480},
abstract = {This article presents a networked wireless telemetry for 384-sensor-pixel neural data acquisition, designed to support high throughput, low power consumption, and meter-scale communication distance. The proposed event-based Address-Event-μPacket (AEμP) encoder with differential binary phase shift keying (DBPSK) modulation addresses the challenging throughput-power trade-off for neural telemetry without compromising transmission range and neural signal fidelity. The encoder compresses data and minimizes protocol overhead as well as memory access. In addition, a duty-cycling IR-UWB transmitter is proposed as communication link for the event-based encoder. To support the multi-access operation, a duty-cycling frequency-locked loop (FLL) is adopted to provide fast-settling packet-level frequency control for frequency division multiplexing (FDM). The FLL achieves warm-start lock within 1 μs with 2.73-MHz RMS variation, meeting the frequency accuracy required for DBPSK at a 125-MHz symbol rate. The measured 3-access-node spectra comply with FCC/ETSI masks in 6-8.5 GHz. The neural signal integrity is validated by evaluating the RMS error between the reconstruction signals and the original in-vivo data captured by a high-pixel count Neuropixels probe. The presented AEμP achieves a 11.52× compression ratio. Fabricated in 65 nm CMOS with active area of 0.12 mm[2], the presented telemetry is capable of transmitting 384 pixels of neural data with low compression error (3.6-μV RMS), at record-low power of 1 mW, and with a 6.6-meter air communication range.},
}

@article {pmid42894183,
year = {2026},
author = {Wu, Y and Dong, Q and Chen, J and Huang, C and Zhang, M},
title = {Temporal synchrony promotes motor recovery and ipsilesional cortical reorganization in BCI-based stroke rehabilitation: a randomized controlled trial.},
journal = {Journal of neuroengineering and rehabilitation},
volume = {23},
number = {1},
pages = {},
pmid = {42894183},
issn = {1743-0003},
support = {XWRCHT20220045//Xuzhou Key medical Talents Project/ ; 2025TD14//New Round of Xuzhou" Pengcheng Talent Program"- High-level Healthcare Talent Recruitment and Development Project/ ; KC25081//Medical and Health General Program of Xuzhou Science and Technology Bureau/ ; },
mesh = {Aged ; Female ; Humans ; Male ; Middle Aged ; *Brain-Computer Interfaces ; Imagination/physiology ; Neuronal Plasticity/physiology ; *Recovery of Function/physiology ; Robotics ; Spectroscopy, Near-Infrared ; Stroke/physiopathology ; *Stroke Rehabilitation/methods ; *Cerebral Cortex/physiology ; },
abstract = {BACKGROUND: The therapeutic mechanisms of brain-computer interface (BCI) rehabilitation remain partially understood. Specifically, it is unclear whether functional gains are driven primarily by active neural engagement or strictly by the temporal contingency between motor intention and sensory feedback. This study aimed to deconstruct the BCI paradigm to isolate the specific contributions of these core components to functional recovery and cortical reorganization in subacute stroke.

METHODS: Seventy-seven subacute stroke patients were randomized into three intervention arms. Group C (synchronous) received robotic assistance strictly contingent on real-time detection of task engagement during motor imagery; Group B (asynchronous) performed identical motor imagery tasks but received temporally uncoupled robotic assistance; and Group A (sham) received sham motor imagery detection and preset passive robotic assistance. The intervention lasted for 3 weeks. The primary clinical outcome was Fugl-Meyer Assessment-Upper Extremity (FMA-UE) scores. Functional near-infrared spectroscopy (fNIRS) was included as an exploratory neuroimaging measure to provide preliminary information on cortical responses associated with the intervention.

RESULTS: Sixty-eight participants were included in the final analysis. After adjustment for baseline values, Group C showed significantly higher post-intervention FMA-UE scores than both Group A (adjusted mean difference = 7.47, 95% CI [1.94, 13.00], P = 0.004) and Group B (adjusted mean difference = 9.72, 95% CI [4.14, 15.30], P < 0.001). Group C also showed significantly higher post-intervention MBI scores than both Group A (adjusted mean difference = 21.15, 95% CI [12.49, 29.81], P < 0.001) and Group B (adjusted mean difference = 21.27, 95% CI [12.42, 30.12], P < 0.001). Exploratory neuroimaging analyses showed that only the change in ipsilesional sensorimotor cortex laterality index during motor imagery was significantly correlated with motor improvement (ρ = 0.404, P = 0.002, q = 0.012), whereas no resting-state or task-related fNIRS group comparisons remained significant after false discovery rate correction.

CONCLUSIONS: Synchronous BCI training was associated with superior clinical improvement compared with asynchronous and sham conditions in subacute stroke. The benefit was observed only when robotic assistance was contingent on real-time task performance, despite all three groups receiving comparable device exposure and motor imagery instruction. The neuroimaging results remain exploratory and should be confirmed in larger studies with prospectively defined imaging endpoints. Trial registration This trial was registered under the Chinese Clinical Trial Registry (ChiCTR2400091197) and was prospectively registered on October 21, 2024. Ethics Approval: Approved by the Ethics Committee of Xuzhou Rehabilitation Hospital (XK-LSW-20241017-009).},
}

Aged
Female
Humans
Male
Middle Aged
*Brain-Computer Interfaces
Imagination/physiology
Neuronal Plasticity/physiology
*Recovery of Function/physiology
Robotics
Spectroscopy, Near-Infrared
Stroke/physiopathology
*Stroke Rehabilitation/methods
*Cerebral Cortex/physiology

@article {pmid42895891,
year = {2026},
author = {Pitt, KM and Sawall, J and Butler, E and Bubak, A and Behboodi, A},
title = {Bridging the Gap Between Design and Access: Toward Visual Scene Displays for Brain-Computer Interface Use in Augmentative and Alternative Communication by Children.},
journal = {Journal of speech, language, and hearing research : JSLHR},
volume = {},
number = {},
pages = {1-19},
doi = {10.1044/2026_JSLHR-26-00267},
pmid = {42895891},
issn = {1558-9102},
abstract = {PURPOSE: Brain-computer interface (BCI) may support communication and computer interaction through neural signals such as the P300 event-related potential (ERP). As a step toward integrating photographic visual scene displays (VSDs) in BCI, this study examined the feasibility of incorporating contextual, naturalistic scenes in which communication items were embedded asymmetrically.

METHOD: To evaluate the effects of contextual scene use, 24 neurotypical children aged 8-13 years completed three experimental conditions: (1) a contextual scene with asymmetrical item placement, (2) an asymmetrical arrangement of the same items without contextual information, and (3) a standard grid. Electroencephalogram data were analyzed for the N170, P300, and N400 ERP components, and offline (predicted) BCI augmentative and alternative communication accuracy was calculated.

RESULTS: The results showed that asymmetry increased frontal and centro-parietal P300 amplitudes, and the scene enhanced occipital N170. Offline BCI performance was similar across the scene and grid conditions, and participants rated the scene as more visually appealing.

CONCLUSION: These findings indicate that asymmetry and contextual embedding within scenes may enhance neural engagement and user experience, providing a preliminary foundation for considering photographic VSDs into pediatric BCI research.},
}

@article {pmid42896470,
year = {2026},
author = {Sun, W and Fu, S and Zhang, J and Zhu, A and Wang, J and Yin, S and Liu, G},
title = {A spatio-temporal attention convolutional network with topology-preserving channel embedding and transformer for EEG decoding.},
journal = {Neural networks : the official journal of the International Neural Network Society},
volume = {206},
number = {Pt A},
pages = {109755},
doi = {10.1016/j.neunet.2026.109755},
pmid = {42896470},
issn = {1879-2782},
abstract = {EEG-based BCI has shown promising applications in motor rehabilitation and emotion recognition. Given the non-stationarity and variability of EEG signals, it is essential to develop classification models that can effectively extract spatio-temporal features while maintaining strong generalization capabilities. This paper proposes a spatio-temporal attention convolutional network for EEG decoding (E-STACNet) to encode the topological features and temporal characteristics of EEG signals within a unified network framework. It consists of channel-attention (CA) block, convolutional (CV) layers, multi-head self-attention (MSA) and classifier layer. Specifically, CA block encodes spatial topological feature via a topology-preserving channel embedding method. Then convolutional layers are employed to encode low-level temporal information and further extract spatial features, followed by MSA block to highlight the most important temporal information of EEG signals. Finally, the outputs are obtained through two fully-connected layers. We conducted experiments on three public EEG-based datasets including BCI Competition IV dataset 2a, Giga DB and DEAP. The results demonstrate E-STACNet achieves average accuracies of 88.00% and 80.62% on datasets 2a and Giga DB, respectively. On the DEAP, it reaches average accuracies of 96.49% and 96.46% for valence and arousal. The proposed model outperforms other state-of-the-art models on all three datasets. Additionally, the ablation study reveals the importance of topology-preserving channel embedding for the proposed network. Visualization analysis shows E-STACNet could obtain the importance of channels at different locations through the channel-attention mechanism, which demonstrates the interpretability of the proposed model. Overall, the accurate and generalization capabilities of E-STACNet will promote the development of EEG-based BCIs.},
}

@article {pmid42896556,
year = {2026},
author = {Kadlec, K and Gamez, J and Aflalo, T and Guan, C and Rosario, ER and Pejsa, K and Bari, A and Pouratian, N and Andersen, RA},
title = {Distinct patterns of whole-body coding in human motor cortex and posterior parietal cortex.},
journal = {Neuron},
volume = {},
number = {},
pages = {},
doi = {10.1016/j.neuron.2026.09.030},
pmid = {42896556},
issn = {1097-4199},
abstract = {Traditional views of cortical motor areas hold that movements of different parts of the body, or effectors, are represented in distinct anatomical locations. The extent of overlap in effector representations has been a topic of debate for decades. We recorded from single neurons in hand areas of the motor cortex (MC) and posterior parietal cortex (PPC) of tetraplegic humans while they attempted movements across the body. We found a population response to every movement tested in both areas; however, the hand knob of MC more selectively responded to hand movements, whereas PPC represented all effectors strongly. Single neurons in MC responded selectively to a single effector or to two effectors that were part of the same limb. In PPC, neurons exhibited mixed selectivity, responding to combinations of effectors. These findings suggest a transition from a highly effector-general representation in PPC to regional effector specificity in MC.},
}

@article {pmid42896567,
year = {2026},
author = {Gorenshtein, A and Adiniaev, Y and Liba, T and Barash, Y and Klang, E and Daniel, O},
title = {The calibration-to-accuracy mapping in P300 spellers does not transport across cohorts and was not restored by tested alignment or local recalibration.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/aeb62c},
pmid = {42896567},
issn = {1741-2552},
abstract = {Objective A calibration-derived score has been related to online P300-speller accuracy in the same session, always within the cohort measured. We evaluated whether a mapping from that score to expected accuracy transports to withheld cohorts. Approach Retrospective secondary analysis of BigP3BCI 1.0.0, the largest evaluation of this relationship to date: 18 of 20 studies contributed 271 participants, 739 session-condition records, and 19,611 character selections. The predictor was a decoder's cross-validated discriminability on calibration epochs, one study withheld from development. Between-cohort variation was summarised by the random-effects standard deviation tau. Main Results The association was positive in all 18 cohorts but varied widely (r 0.190 to 0.928; participant-level r = 0.714, p < 0.001), with 12 CIs excluding zero. Pooled estimation error was 0.098 (95% CI 0.091 to 0.107) against a benchmark of 0.146. Calibration did not transport: intercept tau 0.87 (unrepresented-cohort interval -1.97 to 1.85, log-odds); cohort-specific slopes varied more than tenfold (tau 0.43, interval 0.111 to 2.005). Neither alignment (signal- or score-space) nor either nonlinear boundary meaningfully reduced this heterogeneity. Local recalibration with up to 16 participants did not reliably improve on transporting; refitting both intercept and slope was significantly worse than transporting at the smallest sizes. Median time per selection, a stopping-rule proxy, accounted for 66% of between-cohort slope variance in meta-regression. The four ALS cohorts gave an uncorrected slope spread of 0.223, against 0.560 overall. Significance The evaluated score is a classifier's cross-validated discriminability on each session's calibration data rather than the source studies' undocumented online decoder. That score reproducibly signals within-session accuracy, but its mapping to accuracy is cohort-specific and supports ranking sessions after local validation rather than reporting accuracy elsewhere. The four-cohort ALS subgroup appeared more homogeneous than the full archive, illustrating how conclusions about transportability from a small cohort set can be fragile. .},
}

@article {pmid42896691,
year = {2026},
author = {Zhang, T and Bai, D and Wei, X and Zhuo, L and Zhang, Q and Zhang, J and Jin, Z and Li, L},
title = {Subjective emotional ratings dissociate brain activation and network patterns during counterfactual outcome processing.},
journal = {NeuroImage},
volume = {},
number = {},
pages = {122303},
doi = {10.1016/j.neuroimage.2026.122303},
pmid = {42896691},
issn = {1095-9572},
abstract = {Human decision-making integrates outcome evaluation, counterfactual comparison, and individual traits. We investigated the neural processing of outcome-related emotions during economic decision-making by classifying feedback according to participants' subjective emotional ratings rather than objective gain-loss categories. Participants completed a Wheel of Fortune gambling task during fMRI scanning and provided emotional ratings for feedback outcomes, enabling identification of regret, relief, disappointment, and elation. We also assessed individual differences in regret proneness, maximization tendencies, alternative search, high standards, and decision difficulty. Behaviorally, emotional ratings increased with larger obtained outcomes and decreased as reference outcomes increased, indicating robust modulation of outcome evaluation by counterfactual information Whole-brain analyses revealed greater activation for low-rating than high-rating outcomes, indicating stronger neural responses to negative than positive emotions across feedback types. Complete feedback also elicited greater activation than partial feedback within the same emotional valence. Network-level analyses revealed partially dissociable neural profiles for positive and negative emotions. Positive emotions, including elation and relief, predominantly recruited the somatomotor network, whereas negative emotions, including disappointment and regret, engaged broader distributions across the default mode, dorsal attention, and other large-scale networks. Regret elicited the largest activation extent, whereas elation elicited the smallest. Region-of-interest analyses showed that bilateral orbitofrontal cortex (OFC) responses were significantly greater for negative than positive emotions and for complete than partial feedback conditions. Across all four emotions, right OFC activation was negatively associated with maximization. For elation, relief, and disappointment, right OFC activation was also negatively associated with alternative search and decision difficulty, whereas these associations were absent for regret. Together, these findings demonstrate that subjective emotional evaluation, counterfactual context, and decision-related traits jointly shape the neural processing of decision outcomes, while suggesting relatively trait-independent neural processing of regret.},
}

@article {pmid42898475,
year = {2026},
author = {Fleeting, CR and Gunduz, A and Okun, MS and Wong, JK},
title = {Editorial: Deep brain stimulation think tank: updates in neurotechnology and neuromodulation, volume VI.},
journal = {Frontiers in human neuroscience},
volume = {20},
number = {},
pages = {2007514},
doi = {10.3389/fnhum.2026.2007514},
pmid = {42898475},
issn = {1662-5161},
support = {T32 GM163993/GM/NIGMS NIH HHS/United States ; },
}

@article {pmid42899673,
year = {2026},
author = {Sato, M and Hatakeyama, E and Inoue, M and Kabe, Y and Nobe, S and Yoshida, A and Yamamoto, A and Kita, Y and Shimizu, M and Tomeoka, K and Cohen, MX and Sasai, S},
title = {Delineating neural contributions to electroencephalogram-based speech decoding.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/aeb68d},
pmid = {42899673},
issn = {1741-2552},
abstract = {Objective. Speech Brain-computer interfaces (BCIs) have emerged as a pivotal technology in facilitating communication for individuals with speech impairments. Utilizing electroencephalography (EEG) for noninvasive speech BCIs offers an accessible and affordable solution, potentially benefiting a broader audience. However, EEG-based speech decoding remains controversial especially for overt speech, due to difficulties in separating speech-related neural activities from myoelectric potential artifacts generated during articulation. Moreover, because paradigms for collecting covert speech data are inherently less scalable than those for overt speech, whether the neural dynamics underlying EEG-based decoding of overt speech are also present during covert speech remains an open and practically important question. Approach. Here we aim to delineate the extent of the neural contributions to EEG-based speech decoding, and the degree to which the underlying neural dynamics are shared across overt and covert speech, by employing Explainable AI techniques to a convolutional neural network predicting spoken words based on signals obtained by ultra-high-density (uhd)-EEG. Main results. We found that electrode-wise contributions to the decoding cannot be explained by their mutual information with electromyography (EMG). Furthermore, contributing periods of speech to EEG-based decoding are distinct from those to decoding solely relying on EMG. In contrast, there are significant overlaps in signal timings contributing to EEG-based decoding, regardless of vocal conditions such as overt or covert speech. Notably, the denoising process successfully enhanced the decoding contribution from electrodes within speech-related brain areas for all speech conditions. Significance. Altogether, our findings support the idea that, with appropriate preprocessing, EEG becomes a valuable tool for decoding spoken words based on underlying neural activities. .},
}

@article {pmid42889466,
year = {2026},
author = {Wei, Y and Cui, H and Zhu, J and Chen, C and Qian, Z and Zhang, D and Wang, J and Zhang, T},
title = {An exploration in mental health care: The launch of China's first psychiatric noninvasive brain-computer interface (BCI) and neuromodulation clinic.},
journal = {General psychiatry},
volume = {39},
number = {5},
pages = {e70069},
pmid = {42889466},
issn = {2517-729X},
}

@article {pmid42889891,
year = {2026},
author = {Zhao, Y and Wei, LJ and Sun, XT and Liu, FR and Zhang, L and Zhu, CZ},
title = {Prior photon measurement density functions constructed in standard channel space for functional near-infrared spectroscopy.},
journal = {Biomedical optics express},
volume = {17},
number = {10},
pages = {5034-5047},
pmid = {42889891},
issn = {2156-7085},
abstract = {Photon measurement density function (PMDF) determines the measurement locations and their associated sensitivity to the hemodynamic changes for an fNIRS channel. It is therefore essential for pre-experimental optode arrangement and post-experimental signal analysis and interpretation in fNIRS studies. However, accurate PMDF estimation requires subject-specific MRI data, which are unavailable in common fNIRS studies. Previous studies often rely on PMDFs derived from a single head template (template-PMDF), ignoring inter-subject variability in anatomy and optode placement, and are limited to sparse predefined scalp locations. Here, we propose a methodological framework to construct prior PMDFs (prior-PMDF) by synthesizing PMDFs derived from multiple subjects in a structural MRI database. This approach incorporates inter-subject variability and enables estimation of PMDFs in the standard brain space for arbitrary scalp locations. We demonstrate the utility of the prior-PMDF in estimating regional sensitivities, including voxel-wise sensitivity, sensitivity to intra- and extracerebral tissues, and sensitivity to atlas-defined brain regions. The estimation accuracy of the prior-PMDF is evaluated by comparing it with subject-specific PMDFs derived from MRI data that are not included in the construction database. Furthermore, comparison with template-PMDF showed that the proposed method significantly improves accuracy, thereby supporting more reliable PMDF estimation in MRI-free fNIRS studies.},
}

@article {pmid42890584,
year = {2026},
author = {Tang, B and Wei, J and Liu, Y and He, B and Bao, W and Yang, Y and Wang, W and Hu, W and Duan, A},
title = {Unequal economic impacts of climate change via idealized carbon dioxide removal: Insights from deep learning.},
journal = {Innovation (Cambridge (Mass.))},
volume = {7},
number = {10},
pages = {101505},
pmid = {42890584},
issn = {2666-6758},
abstract = {Rapidly reducing carbon dioxide (CO2) levels is essential to meeting the Paris Agreement's temperature targets. Previous assessments of CO2 removal (CDR) have primarily focused on the hysteresis and invertibility of climate change itself, overlooking quantitative analysis of potential economic impacts of climate change via CDR. In this study, we first develop a powerful neural network model, EconClimNet, trained on decades of economic data from 1,554 sub-national regions worldwide and 111 climate indices derived from the fifth-generation ECMWF atmospheric reanalysis (ERA5). Compared to popular machine learning algorithms used in the Earth science community, EconClimNet achieves superior performance in capturing the intricate relationship between climate indices and economic outcomes. On this basis, we apply EconClimNet to the outputs from idealized CO2 ramp-up and ramp-down experiments from phase 6 of the Coupled Model Intercomparison Project (CMIP6). It is found that altering climate trajectories through idealized CDR can yield significant long-term economic benefits globally, with Oceania overall positioned to experience the greatest gains with an increasing trend of about 1.41 [0.73-8.76] million USD year[-1]. Unequal economic effects also emerge across income levels. Rich countries see more benefits (an increasing trend of about 1.56 [0.31-6.21] million USD year[-1]) via CDR, mainly contributed by services (an increasing trend of about 0.03 [0.00-0.23] million USD year[-1]). Our findings underscore the need for governments to foster the innovation, development, and prudent deployment of CDR technologies, as well as to advance global CDR strategies that prioritize vulnerable nations from a climate justice perspective.},
}

@article {pmid42882312,
year = {2026},
author = {Li, Y and Yang, R and Xu, X and Jin, L and Zhang, M and Chen, L},
title = {Lower Task-Estimated Visual Working Memory Capacity in Adults with Epilepsy: A Cross-Sectional Quantitative Assessment Using Eye Tracking.},
journal = {Diagnostics (Basel, Switzerland)},
volume = {16},
number = {19},
pages = {},
pmid = {42882312},
issn = {2075-4418},
support = {2024ZDZX0018//Science and Technology Department of Sichuan Province/ ; },
abstract = {Background/Objectives: Cognitive dysfunction is clinically relevant in epilepsy, and objective task-based measures could complement neuropsychological assessment, but visual working memory across graded loads and task-embedded gaze remain incompletely characterized. We quantified task-estimated capacity and performance using gaze-triggered change localization. Methods: We analyzed data from 222 adults with epilepsy, including focal and generalized epilepsy, and 133 healthy controls. Participants identified the changed item in a test array by sustained gaze; set size (2-7) followed a staircase procedure. Outcomes were three task-performance measures, two capacity-related measures, and eight eye-movement measures. Repeated-measure outcomes were analyzed with mixed-effect models and the capacity-related measures by ordinary least squares, with Benjamini-Hochberg correction. Results: Adults with epilepsy had lower mean set size and K at set size 4 (K4), lower accuracy, longer response time, and greater final gaze-to-target distance than controls (all q < 0.05; absolute Cohen's d, 0.20-0.45). K4 was 2.66 versus 3.03 items. Differences persisted after adjustment for educational attainment and were directionally similar in focal and generalized epilepsy. No interaction between group and set size survived correction for task performance; group differences did not vary detectably across the observed load range. No eye-movement group difference survived correction. Conclusions: In a relatively large adult cohort, task-estimated visual working memory capacity and task performance were lower in epilepsy, without broad corrected group differences in the selected array-level eye-movement measures. By combining graded load with gaze-based response registration, the study characterizes the magnitude, load dependence, timing, and spatial precision of visual working memory performance differences within a single task.},
}

@article {pmid42884285,
year = {2026},
author = {Lee, CH and Huang, CM},
title = {Gradient-Free Filter Importance for Structured Pruning of Compact EEG Networks.},
journal = {Sensors (Basel, Switzerland)},
volume = {26},
number = {19},
pages = {},
pmid = {42884285},
issn = {1424-8220},
support = {NSTC-115-2221-E-324-016-.//National Science and Technology council/ ; },
mesh = {*Electroencephalography/methods ; Brain-Computer Interfaces ; Humans ; Algorithms ; *Signal Processing, Computer-Assisted ; },
abstract = {Deploying EEG classification networks on resource-constrained brain-computer interface (BCI) hardware motivates structured filter pruning, yet which importance criterion to trust in already-compact architectures is unclear. We present a leakage-free, physically materialized comparison of five filter-importance criteria-ablation-based structural sensitivity analysis (SSA), first-order Taylor, L1-norm, geometric median, and random-across four datasets, two paradigms (motor imagery and ERP), two architectures, multiple capacities, and with and without fine-tuning. On EEGNet, after fine-tuning, SSA and Taylor are statistically equivalent: pooled over the four datasets their accuracy difference falls within a ±1 percentage-point margin (TOST, p=0.01), and on ShallowConvNet SSA is never worse than Taylor; both criteria significantly outperform the weight-based and random ones. SSA's value is therefore not a more accurate ranking but its formulation: on a weight-quantized (8-bit) EEGNet, where the loss gradient-and hence Taylor-is undefined on the quantized graph used here, SSA is the best gradient-free criterion, significantly beating L1, geometric-median, and random pruning in aggregate-extending structured pruning to the quantized, gradient-free but editable deployment regime. We further show that the pruning schedule and fine-tuning matter more than the choice between strong criteria, and that training large then pruning with fine-tuning recovers accuracy equivalent to a directly trained smaller model. The results clarify where the importance criterion does and does not matter for compact EEG decoders.},
}

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

@article {pmid42884399,
year = {2026},
author = {Habibzadeh, H and Zois, DS and Norton, JJS},
title = {MetaIDNet: A Fast Brain-Computer Interface for Color Vision Assessment.},
journal = {Sensors (Basel, Switzerland)},
volume = {26},
number = {19},
pages = {},
pmid = {42884399},
issn = {1424-8220},
support = {1R21EB036221-02/NH/NIH HHS/United States ; 1P41EB018783-10/NH/NIH HHS/United States ; },
mesh = {Humans ; *Brain-Computer Interfaces ; *Color Vision/physiology ; Evoked Potentials, Visual/physiology ; Neural Networks, Computer ; Electroencephalography ; Photic Stimulation ; Male ; Algorithms ; },
abstract = {Our laboratory has recently demonstrated objective, quantitative, and precise brain-computer interface (BCI)-based color vision (CV) assessment. During assessment, participants look at a flashing stimulus that alternates between two light sources. Light sources that appear to be different colors elicit larger steady-state visual evoked potentials (SSVEPs); those that appear to be similar colors elicit smaller SSVEPs. Thus, the stimulus-evoked responses depend on an individual's CV. BCI-based CV assessments evaluate CV by observing SSVEP sizes elicited over a color space; their accuracy depends on the sampling strategy and noise in the individual observations. Existing systems uniformly sample the color space and average multiple SSVEPs observed from the same location to reduce noise. Effectively, these systems collect additional observations to improve assessment accuracy, but at the expense of assessment speed. We present MetaIDNet, a closed-loop BCI that uses Gaussian process regression (GPR) and a neural network to improve the speed-accuracy tradeoff of BCI-based CV assessment. We perform offline and online experiments with ten and eight participants, respectively, to evaluate the speed and accuracy of MetaIDNet and compare it with two existing methods. Our offline experiments show that MetaIDNet improves CV assessment speed by 70% (while maintaining accuracy), and our online experiments show that MetaIDNet improves CV assessment accuracy by 40% (while maintaining the same speed). The results confirm that MetaIDNet significantly improves the speed and accuracy of BCIs for CV assessment.},
}

Humans
*Brain-Computer Interfaces
*Color Vision/physiology
Evoked Potentials, Visual/physiology
Neural Networks, Computer
Electroencephalography
Photic Stimulation
Male
Algorithms

@article {pmid42884552,
year = {2026},
author = {Serna, B and Salazar, R and Alonso-Silverio, GA and Baltazar, R and Ventura-Molina, E and Alarcón-Paredes, A},
title = {Fear-EEG-AI: A Conceptual and Operational Framework for Designing EEG-AI Systems for the Inference of Fear-Compatible Neurophysiological States.},
journal = {Sensors (Basel, Switzerland)},
volume = {26},
number = {19},
pages = {},
pmid = {42884552},
issn = {1424-8220},
mesh = {*Electroencephalography/methods ; Humans ; *Fear/physiology ; *Artificial Intelligence ; },
abstract = {Electroencephalography (EEG) and artificial intelligence (AI) are central tools in affective computing, yet fear-related inference remains constrained by a gap between experimental performance and deployable evidence: high accuracy on controlled datasets guarantees neither neurobiological specificity, nor robustness across users, nor traceable inference. EEG does not measure fear directly; it captures cortical dynamics compatible with threat processing, arousal or vigilance. This paper introduces the Fear-EEG-AIFramework, a technology-independent architecture for designing, validating and deploying EEG-AI systems targeting fear-compatible neurophysiological states. Building on a prior systematic review, it organizes development into three hierarchical levels-neurobiological foundation, neuroelectric correlates and computational modeling-governed by six cross-cutting principles, and makes them evaluable through eight observable dimensions, the Fear-EEG-AI Alignment Index (FAI) and five validation evidence levels. Applied to the eight eligible studies of that corpus by two raters who had not drafted the instrument, with adjudication by a third, the rubric reached 67.2% exact agreement (quadratic-weighted κ=0.741; ICC(2,1) =0.744). Under 10,000 admissible weightings the equal-weight ordering survived in 10.1% of draws, and one of 56 ordered pairs met a stability criterion fixed before the pairwise results were tabulated. The index is therefore reported as a descriptor of methodological documentation and not as a ranking of implementations.},
}

*Electroencephalography/methods
Humans
*Fear/physiology
*Artificial Intelligence

@article {pmid42886472,
year = {2026},
author = {Anniwa, D and Chen, Z and Chen, X and Yuan, W and Lang, H and Chen, N and Zhang, N and Qin, M},
title = {Advanced biomaterial technologies for neural regeneration in central nervous system diseases: mechanisms and applications.},
journal = {Nanoscale},
volume = {},
number = {},
pages = {},
doi = {10.1039/d6nr02348g},
pmid = {42886472},
issn = {2040-3372},
abstract = {Effective repair of central nervous system (CNS) injuries remains highly challenging due to several formidable barriers, including severe structural disruption, hostile microenvironments, and limited intrinsic regenerative capacity. In response, biomaterials incorporated with therapeutic agents provide advanced interventional strategies for neural regeneration. This review systematically summarizes recent progress in biomaterial-based CNS repair, focusing on three core strategies: physical remodeling of the neural structure, targeted drug delivery systems for biochemical regulation, and cellular intervention. Furthermore, we critically analyze the manufacturing and regulatory bottlenecks currently impeding clinical translation. Finally, we propose the integration of neural organoids and brain-computer interface (BCI) technologies as a promising trajectory for next-generation intelligent neural repair materials.},
}

@article {pmid42887837,
year = {2026},
author = {Duwadi, S and Rogers, D and Boyd, AD and Carlton, LB and Zhang, Y and Gaona, AK and Pathiyaparambath, AD and Chaudhury, R and Magee, G and Zimmermann, BB and O'Brien, WJ and von Lühmann, A and Boas, DA and Yücel, MA and Sen, K},
title = {Decoding Spatial Attention in the Cocktail Party Problem Using Whole-head High-Density fNIRS.},
journal = {NeuroImage},
volume = {},
number = {},
pages = {122292},
doi = {10.1016/j.neuroimage.2026.122292},
pmid = {42887837},
issn = {1095-9572},
abstract = {Spatial attention is critical for solving the cocktail party problem, a longstanding problem in neuroscience and artificial speech recognition. The ability to decode where humans are attending in a cocktail party like scene would empower applications in brain computer interfaces and assistive devices such as hearing aids. Here we demonstrate that, in an overt attention task, the attended spatial location can be decoded robustly from single trial hemodynamic responses, using a whole head high density fNIRS system. We also identify critical brain regions that make the highest contribution to decoding accuracy. Specifically, we find that decoding based on a small fraction of channels located within the left and right inferior parietal lobule (IPL) achieves maximal decoding accuracy comparable to all channels. Decoding fell close to chance in a covert auditory condition in which participants held central fixation while attending to the same talkers, indicating that auditory attention alone without orientation is not sufficient for decoding. Finally, control experiments suggest that overt-condition decoding is influenced by sustained eye position/gaze direction, not eye movement. These results support the development of fNIRS-integrated BCIs and assistive devices controlled by overt spatial attention.},
}

@article {pmid42881397,
year = {2026},
author = {Chen, Y and Gao, Y and Zhou, J and Zhang, Z and Zong, J and Liu, Y},
title = {Brain-computer interfaces: innovations and challenges in the field of cerebral palsy rehabilitation.},
journal = {Italian journal of pediatrics},
volume = {52},
number = {1},
pages = {},
pmid = {42881397},
issn = {1824-7288},
support = {Dalian Association of Science and Technology [2020]No.42//Dalian Association of Science and Technology Expert Workstation Project/ ; },
mesh = {Humans ; *Cerebral Palsy/rehabilitation ; *Brain-Computer Interfaces ; Child ; Activities of Daily Living ; },
abstract = {Brain-computer interfaces have shown great potential in the rehabilitation of Cerebral palsy, improving the ability of children with cerebral palsy to perform activities of daily living by replacing and restoring lost neurological functions, as well as accurately evaluating cognitive functions and facilitating rehabilitation in conjunction with various rehabilitation techniques. Nonetheless, numerous challenges persist in the application of brain-computer interfaces for cerebral palsy. Future research should focus on developing enhanced and more dependable brain-computer interfaces, advancing the hardware, and establishing ethical and regulatory frameworks to facilitate their utilisation in cerebral palsy rehabilitation.},
}

Humans
*Cerebral Palsy/rehabilitation
*Brain-Computer Interfaces
Child
Activities of Daily Living

@article {pmid42877582,
year = {2025},
author = {Chen, ZS and Lu, BL and Wu, W},
title = {Representation Learning and Foundation Models for Electroencephalography Analyses: Current trends, fundamental insights, and future directions?.},
journal = {IEEE signal processing magazine},
volume = {42},
number = {5},
pages = {36-57},
pmid = {42877582},
issn = {1053-5888},
support = {UG3 NS135170/NS/NINDS NIH HHS/United States ; RF1 NS121776/NS/NINDS NIH HHS/United States ; R01 MH118928/MH/NIMH NIH HHS/United States ; P50 MH132642/MH/NIMH NIH HHS/United States ; RF1 DA056394/DA/NIDA NIH HHS/United States ; R01 MH139352/MH/NIMH NIH HHS/United States ; R01 NS121776/NS/NINDS NIH HHS/United States ; R01 NS123928/NS/NINDS NIH HHS/United States ; },
abstract = {Since the first ground-breaking human electroencephalography (EEG) recordings in 1924 [1], the past century has witnessed a tremendous growth of EEG applications in cognitive neuroscience, clinical and engineering applications due to EEG's low operational cost and mobility [2]. On the one hand, advances in high-density non-invasive scalp EEG or invasive intracranial EEG (iEEG) have offered both excellent temporal resolution and increasingly improved spatial resolution to study brain functions and their link to emotions, memory, learning, and diseases. EEG-based brain-computer interface (BCI) can offer new dimensions for entertainment, virtual reality, neurofeedback and closed-loop therapy. On the other hand, recent advances in artificial intelligence (AI) and machine learning have opened new opportunities for analyses of EEG and other neural data [3]. This article aims at presenting an overview of the cutting-edge machine learning techniques for EEG analyses. By leveraging large-scale EEG data with state-of-the-art representation learning and transfer learning paradigms, we are empowered to discover latent EEG features that are proved useful for clinical care and BCI. We discuss some general principles of representation learning and show walk-through practical examples of EEG analysis. The article also aims at highlighting the effort of applying AI models to discover neuroscience insights and linking them to the fundamentals of EEG signal analyses from a signal-processing perspective. While our focus is on EEG and iEEG signals, most of the approaches discussed here are generally applicable to other brain signal modalities.},
}

@article {pmid42878278,
year = {2026},
author = {Xie, Y and Deng, J and Wei, C and Zhao, J and Lan, L and Kuang, Y and Cheng, M and Liu, C and Tian, Y and Wang, H and Yue, W and Liu, J and Ni, Z and Wang, Z and Li, M},
title = {Scalable Patterning of Organic Electrochemical Transistors for Logic Gates and Neuromorphic Computing.},
journal = {Small (Weinheim an der Bergstrasse, Germany)},
volume = {},
number = {},
pages = {e76279},
doi = {10.1002/smll.76279},
pmid = {42878278},
issn = {1613-6829},
support = {2025YFF0518300//National Key R&D Program of China/ ; //Self-deployment Project of State Key Laboratory of Fabrication Technologies for Integrated Circuits/ ; YSBR-146//CAS Project for Young Scientists in Basic Research/ ; //Fundamental Research Funds for the Central Universities/ ; JQ24030//Beijing Natural Science Foundation/ ; },
abstract = {Organic electrochemical transistors (OECTs) are promising building blocks for bioelectronics owing to their efficient ion-electron coupling, soft-material compatibility and high transconductance. However, their large off-state currents and poor scalability have so far limited high-density integration and circuit-level functionality. Here we introduce a patterning strategy suitable for solution‑processable organic semiconductors, enabling high‑fidelity OECT arrays with suppressed leakage and off‑state currents, on/off ratios up to 10[7], and good device‑to‑device uniformity. Leveraging these patterned OECTs, we demonstrate the integration of logic circuits, including complementary inverters as well as NAND and NOR gates, exhibiting high voltage gain (up to 203 V/V), large noise margins (92.17%) and stable operation at low supply voltages. Benefiting from the patterning approach and the AlOx-induced suppression of off-state current, the static power consumption of the inverter is reduced by approximately 90%. Beyond digital logic, we further show that these devices emulate key synaptic functions such as short-term plasticity and enable reservoir computing tasks, defining a manufacturable platform for integrated bioelectronic circuitry and neuromorphic architectures. This patterning strategy therefore provides a general route toward high-density biointerfaces and next-generation brain-computer interfaces.},
}

@article {pmid42879210,
year = {2026},
author = {Hu, AQ and Mao, ZX and Li, YY and Chen, GD and Chen, T and Deng, RH and Zhang, YY and Zhang, ZY and Huang, S and Dai, WJ and Xiao, ZC and Yang, X and Tao, YJ and Yi, LL and Wu, JL and Huang, YQ and Wang, HY and Li, T and Guo, WJ},
title = {Unveiling core clinical features of self-harm: A network and mediation analysis study among Chinese psychiatric adolescents.},
journal = {Journal of behavioral addictions},
volume = {},
number = {},
pages = {},
doi = {10.1556/2006.2025.00574},
pmid = {42879210},
issn = {2063-5303},
abstract = {BACKGROUND AND AIMS: Self-harm imposes a substantial burden among adolescents, yet targeted interventions remain limited. This study examined the interrelationships among clinically relevant characteristics of self-harm outcomes, non-suicidal motives (NSMs), addictive features (AF), self-injury methods (SIMs), and suicidality (SUI), to identify core clinical features for more precise intervention.

METHODS: A total of 547 adolescent patients (13-19 years) who had engaged in self-harm during the past year were recruited from two Chinese hospitals. Self-report questionnaires assessed NSMs, AF, SUI, frequencies of non-suicidal self-injury (NSSI) and suicidal self-injury (SSI), and SIMs. Exploratory factor analysis (EFA) was used as a data-driven approach to identify patterns of co-occurrence among SIMs. Network and exploratory mediation analyses were subsequently conducted to evaluate statistical interrelationships among clinical features.

RESULTS: Among participants, 69.8% engaged in both NSSI and SSI. EFA identified two empirically derived groupings of SIMs: self-injury using external tools (SIET) and self-injury involving direct bodily actions (SIDBA). Network analysis identified AF as the strongest bridge symptom and SIET with the highest node strength. Exploratory mediation analyses suggested that AF statistically accounted for substantial proportions of the associations between NSMs and multiple self-harm related outcomes, including NSSI (41.3%), SSI (75.0%), SIET (62.3%), SIDBA (45.4%), and SUI (75.7%).

DISCUSSION AND CONCLUSIONS: AF and SIET emerged as important clinically relevant features of self-harm. Greater attention to these features may help improve clinical assessment and targeted intervention for adolescents with self-harm.},
}

@article {pmid42880317,
year = {2026},
author = {Natarajan, AM and Sajda, P},
title = {Peripheral physiological decoding of arousal in a boundary-avoidance task: signal validity and individualized control.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/aeb4ec},
pmid = {42880317},
issn = {1741-2552},
abstract = {Scoring well on a task-demand label does not show that a decoder measures arousal. We ask whether peripheral physiology decodes demand better than EEG, whether the index behaves like graded arousal rather than a clock, and what a controller could do with it. Approach. We re-analysed a public boundary-avoidance flight task in which sixteen participants received EEG neurofeedback, half-sham feedback or silence. A decoder trained on peripheral signals (cardiac, respiratory, electrodermal, pupillometric) from an open-loop calibration block was compared with a retrained EEG decoder, filter-bank common spatial patterns, and motor-only and shuffled-label controls. All analyses are offline; the decoder was never in the loop. Main results. The peripheral decoder reached 93.0% within-subject AUC against 85.2% for EEG. Ring size shrinks on a fixed schedule, so accuracy alone cannot separate arousal from trial position. Three tests did: 48.6% of hard-course trials ended below the scale midpoint; the index separated the courses at matched elapsed time (p= 1.4 ×10-31), where EEG did not; and it tracked the autonomic channels. Only the peripheral index recovered an inverted-U with performance, robust in a duration-safe hazard analysis (p= 3.5 ×10-10). Arousal rose 14.1 points in the 10 s before failure; sustained between-condition differences above 7.5 points are excluded. Subjects varied reliably in how strongly excess arousal predicted failure, predictably from pre-loop calibration. Personalised control bands gave the best held-out discrimination, though at N = 16 indistinguishable from a retuned universal band. Significance. Arousal is a state of the whole body, and here the body reads it better than the scalp. The results separate three often-conflated requirements: accuracy on a demand label, validity as an arousal signal, and usefulness for control. The signal that added information over a fixed rule was distance above a band set by each subject's recent variability.},
}

@article {pmid42880669,
year = {2026},
author = {Chrisilla, S and Selvakumari, RS},
title = {A GRAPH CONVOLUTION BASED DUAL-BRANCH NETWORK FOR DECODING OF UPPER-LIMB MOTOR IMAGERY EEG SIGNALS.},
journal = {Journal of neuroscience methods},
volume = {},
number = {},
pages = {110921},
doi = {10.1016/j.jneumeth.2026.110921},
pmid = {42880669},
issn = {1872-678X},
abstract = {BACKGROUND: Analysing motor imagery EEG signals using conventional Euclidean feature computation is challenging due to low spatial resolution, non-stationarity, and complex inter-channel dependencies, which limit robust subject-independent generalization in brain-computer interface systems.

NEW METHOD: This study proposes a graph-based convolutional framework for subject-independent decoding of upper-limb motor imagery. The model integrates graph convolution with skip connections and handcrafted time-frequency features, together with parallel spatial and temporal convolutional pathways. Graph convolution captures functional connectivity between electrodes, while convolutional modules enhance spatial-temporal representation learning. Subject-independent evaluation is ensured using Stratified Group K-Fold (SGKF) and Leave-One-Subject-Out (LOSO) validation protocols.

RESULTS: The framework was evaluated on the PhysioNet, BCI IV-2a, and BCI IV-2b datasets using multiple electrode configurations. For the PhysioNet dataset, three upper-limb classes were considered, achieving SGKF accuracies of 86.42%, 91.01%, and 93.91% using 64, 21, and 5 electrodes, respectively, and LOSO accuracies of 92.25%, 95.72%, and 96.01%. On the BCI IV-2a dataset, SGKF accuracies of 86.13%, 89.01%, and 92.33% were obtained, with LOSO accuracies reaching 97.12% using five electrodes. For the BCI IV-2b dataset, accuracies of 88.60% (SGKF) and 93.35% (LOSO) were achieved using only three electrodes.

The proposed graph-based framework demonstrates competitive performance to representative graph-based methods while requiring fewer trainable parameters and EEG electrodes.

CONCLUSION: These findings demonstrate that graph-based deep learning enables accurate, subject-independent motor imagery decoding with reduced model complexity and minimal electrode usage, supporting practical and resource-efficient brain-computer interface systems.},
}

@article {pmid42585091,
year = {2026},
author = {Yang, R and Xu, S and Pang, W and Zhang, Y},
title = {Perception-Feedback Mapping Strategies for Flexible Human-Machine Interfaces.},
journal = {Advanced materials (Deerfield Beach, Fla.)},
volume = {38},
number = {58},
pages = {e74589},
doi = {10.1002/adma.74589},
pmid = {42585091},
issn = {1521-4095},
support = {L242069//Beijing Natural Science Foundation/ ; 12225206//National Natural Science Foundation of China/ ; 12402199//National Natural Science Foundation of China/ ; 2025Z11DSB001//Tsinghua Initiative Scientific Research Program/ ; 2024YFB4707300//National Key R&D Program of China/ ; //New Cornerstone Science Foundation/ ; //Tsinghua National Laboratory for Information Science and Technology/ ; },
mesh = {Humans ; *Feedback ; *Man-Machine Systems ; *Perception ; Brain-Computer Interfaces ; },
abstract = {The field of flexible human-machine interfaces (HMIs) has been evolving rapidly with progress in materials science, structural design, device fabrication, and system integration. Developments of advanced flexible sensing and feedback devices have remarkably expanded the perceptual modalities, feedback capabilities, and functional scope of these systems. However, the broader deployment of flexible HMIs depends not only on the device performance of sensing/feedback components, but also on the control strategy that translates sensed signals into command outputs for feedback modules. This review highlights recent progress in such perception-feedback mapping strategies for flexible HMIs. Representative sensing and feedback modules, together with common system-level physical architectures, are briefly introduced as a basis for analyzing the relationship between sensed inputs and feedback outputs. A systematic framework is then proposed based on the principle of signal flow from the perception to the feedback side, allowing perception-feedback mapping strategies to be classified into command-based mapping, continuous mapping, and closed-loop control. Comprehensive discussions on these three categories of mapping strategies are presented, covering their operating principles, system implementation, and practical applications. Finally, we discuss current challenges and future directions toward more robust, generalizable, and integrated flexible HMI systems.},
}

Humans
*Feedback
*Man-Machine Systems
*Perception
Brain-Computer Interfaces

@article {pmid42869941,
year = {2026},
author = {Sun, C and Bao, K and Sun, B and Tang, B and Yang, X and Chen, Z and Tang, Y and Deng, Q and Wu, Y and Liu, L and Tang, R and Li, Z and Lei, K and Shi, Y and Shieh, W and Qiu, M and Lin, H and Li, L},
title = {Geometrical-optics resonator photonics.},
journal = {Nature communications},
volume = {17},
number = {1},
pages = {},
pmid = {42869941},
issn = {2041-1723},
abstract = {Conventional waveguide optical resonators face inherent limitations of low damage thresholds, limited coupling tunability, and stochastic resonance. To overcome these, we introduce a geometrical-optics resonator (GOR) framework, emulating free-space reflection and thin-film coupling on-chip. Using ultra-wide waveguides with total internal reflection (TIR) mirrors and frustrated-TIR couplers, GORs eliminate conventional constraints. Fabricated devices achieve intrinsic Q-factors >1 million and enable tunable microwave photonic filters operating distortion-free at about 10× higher optical power. Design flexibility allows precisely tuned coupled resonators, facilitating cascaded networks for spectral engineering. Demonstrated GOR photonic molecules exhibit dynamically adjustable mode splitting (0.49 free spectral range), nearing theoretical limits. The platform also realizes record-large-scale Floquet topological photonic insulators with 1281 GOR lattices, supporting robust edge modes against defects and a broadband continuum. The GOR sets new performance benchmarks, enabling scalable platforms for quantum simulation and topological networks while driving integrated photonics toward higher density and system-level integration.},
}

@article {pmid42870375,
year = {2026},
author = {Cao, R and Miao, M and Gan, Y},
title = {The interplay of family support and family incivility on adolescent mental health.},
journal = {Journal of research on adolescence : the official journal of the Society for Research on Adolescence},
volume = {36},
number = {4},
pages = {e70290},
doi = {10.1111/jora.70290},
pmid = {42870375},
issn = {1532-7795},
support = {2024ZD09//Peking University Health Science Center Medical Education Research Funding Project/ ; CECA21105//Young Scholar Project of Beijing Education Sciences Planning Program during the 14th Five-Year Plan/ ; },
mesh = {Humans ; Adolescent ; Female ; Family Support ; Male ; *Mental Health ; Depression/psychology ; *Incivility ; Anxiety/psychology ; Longitudinal Studies ; China ; Personal Satisfaction ; *Family Relations/psychology ; },
abstract = {Both positive and negative family interactions are closely associated with adolescent mental health. Using an integrated point of view, this study used a longitudinal design to examine the interplay patterns of family support and family incivility on adolescent mental health. Two waves of data were collected from 999 Chinese high school students. Polynomial regressions and response surface analyses were used to examine interplay patterns of family support and incivility, specifically, their congruence effect and incongruence effect on mental health outcomes, including well-being (life satisfaction and meaning in life) and psychopathological symptoms (anxiety and depressive symptoms). Model comparison was conducted within a series of models to select the best fitting model for each mental health outcome. As for positive mental health outcomes, both life satisfaction and meaning in life had a curvilinear correlation with the incongruence of family support and family incivility. The correlations were larger when family support was higher than family incivility than vice versa. Additionally, a significant interaction effect of family support and family incivility was found on meaning in life. Specifically, the positive association between family support and meaning in life was weaker when family incivility was high. As for negative mental health outcomes, both anxiety and depressive symptoms were linearly correlated with the incongruence of family support and family incivility. Specifically, greater family support relative to family incivility was associated with lower levels of anxiety and depressive symptoms. These findings promoted understanding of the impact of family environment on adolescent mental health and highlighted the importance of considering both positive and negative forms of family interaction.},
}

Humans
Adolescent
Female
Family Support
Male
*Mental Health
Depression/psychology
*Incivility
Anxiety/psychology
Longitudinal Studies
China
Personal Satisfaction
*Family Relations/psychology

@article {pmid42870992,
year = {2026},
author = {Zaway, L and Zaway, I and Ktari, J and Amor, NB and Delahoche, L and Alaoui, LC and Jallouli, M},
title = {Optimized hybrid Transformer-based model for EEG classification: application to wheelchair control.},
journal = {AIMS neuroscience},
volume = {13},
number = {3},
pages = {452-475},
pmid = {42870992},
issn = {2373-7972},
abstract = {The purpose of this study is to improve the classification of electroencephalogram (EEG) signals for motor imagery-based brain-computer interfaces (MI-BCIs), thus enabling more reliable user interactions in assistive technologies such as autonomous wheelchairs. The study contributes by systematically investigating and comparing the performance of both simple (Transformer, Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM)) and hybrid deep learning architectures (Transformer + CNN, Transformer + LSTM, CNN + LSTM) for EEG signal classification using multiple evaluation metrics, including accuracy, ROC AUC, confusion matrices, and prediction time. The findings demonstrate that the Transformer + CNN hybrid model consistently outperforms all other architectures, thereby achieving perfect classification accuracy (1.00), zero misclassifications, flawless AUC scores (1.000), and a competitive inference time (1.17 s), thus making it highly suitable for real-time MI-BCI applications. In contrast, simple models, particularly LSTM, show lower accuracies and reduced stabilities. Overall, this study highlights the advantage of combining attention mechanisms and convolutional architectures to effectively capture both spatial and contextual patterns in nonstationary EEG signals, thus establishing Transformer-based hybrid models as a promising approach for robust, real-time EEG classifications in next-generation MI-BCI systems.},
}

@article {pmid42871320,
year = {2026},
author = {Wang, S and Wu, Q and Lei, Q and Li, C and Lei, F},
title = {Brain-computer interfaces in pediatric populations: research advances, application challenges, and future directions.},
journal = {Frontiers in pediatrics},
volume = {14},
number = {},
pages = {1936801},
pmid = {42871320},
issn = {2296-2360},
abstract = {Brain-computer interfaces (BCIs) establish direct communication pathways between the brain and external devices. Leveraging the unique neuroplasticity of the developing brain, BCIs have emerged as a novel rehabilitative technology for children with neurodevelopmental disorders, including cerebral palsy, autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), and perinatal brain injury. This review comprehensively examines five major BCI paradigms-P300, motor imagery (MI), steady-state visual evoked potentials (SSVEP), functional near-infrared spectroscopy (fNIRS), and multimodal hybrid systems-with respect to their pediatric adaptability, technical limitations, and optimization strategies. We summarize the existing evidence across four key clinical applications: augmentative and alternative communication (AAC), motor function rehabilitation, ADHD attention training, and ASD social-emotional intervention. Key intrinsic (cognitive and physiological) and extrinsic (device and environmental) variables affecting BCI performance in children are analyzed, along with general optimization approaches including personalized parameter tuning, pre-training neurofeedback, and lightweight hardware modifications. Furthermore, we examine multilayer barriers to pediatric BCI adoption, including user experience, clinical workforce readiness, translational pathways, and child-specific ethical considerations. Preliminary evidence suggests that non-invasive BCIs can be feasibly used in children above 5 years of age, although this finding is based on short-term studies with small sample sizes and requires further validation through larger longitudinal trials. However, the immature pediatric brain, short attention spans, and low device tolerance preclude direct translation of adult BCI systems. While evidence is more established for AAC and motor rehabilitation in cerebral palsy, large-scale long-term data are still lacking for ADHD and ASD interventions. Future efforts should prioritize child-specific lightweight wearable devices, multicenter longitudinal trials, immersive integration with virtual reality (VR) and social robots, standardized clinical service frameworks, and ethical guidelines for invasive pediatric BCIs, thereby facilitating the transition of BCI technology from laboratory to clinical and home settings and maximizing its value in promoting neurodevelopmental plasticity in children with special needs.},
}

@article {pmid42874874,
year = {2026},
author = {Pan, H and Ding, P and Wang, F and Ma, Y and Gong, A and Nan, W and Li, N and Fu, Y},
title = {Post-Traumatic Stress Disorder Recognition Using the fNIRS Dual-Attention Network.},
journal = {Journal of neuroscience methods},
volume = {},
number = {},
pages = {110925},
doi = {10.1016/j.jneumeth.2026.110925},
pmid = {42874874},
issn = {1872-678X},
abstract = {BACKGROUND: Post-traumatic stress disorder (PTSD) is a prevalent psychological disorder that can impair daily functioning and mental health. Current clinical assessment mainly relies on structured interviews and symptom scales, which may introduce subjectivity and inter-rater variability. Functional near-infrared spectroscopy (fNIRS) provides a noninvasive approach for measuring task-evoked hemodynamic responses and may offer objective physiological information for auxiliary PTSD recognition.

NEW METHOD: This study proposes an end-to-end deep learning framework, the fNIRS Dual Attention Network (FDANet), for PTSD recognition. Ten PTSD patients and ten healthy controls (HC) were recruited, and prefrontal hemodynamic signals, including oxyhemoglobin (HbO), deoxyhemoglobin (HbR), and total hemoglobin (HbT), were recorded during fear and neutral facial-expression stimuli. The three hemodynamic signals were concatenated along the channel dimension to form the model input. FDANet combines Transformer-based temporal modeling, one-dimensional temporal-channel attention, and convolutional feature extraction to learn discriminative representations from multichannel fNIRS signals.

RESULTS: The model was evaluated using subject-wise nested cross-validation with five outer folds. At the trial level, FDANet achieved an accuracy of 79.4%. At the subject level, FDANet achieved an accuracy of 95.0%.

Compared with the evaluated traditional machine-learning and deep-learning baseline methods, FDANet achieved higher classification accuracy at both trial and subject levels.

CONCLUSIONS: These findings indicate that task-evoked prefrontal fNIRS signals contain discriminative information for PTSD recognition. Given the portability and noninvasive measurement characteristics of fNIRS, FDANet may support the future development of objective auxiliary screening tools for PTSD, although further validation with larger and independent samples is required.},
}

@article {pmid42876598,
year = {2026},
author = {Schilling, A and Krauss, P},
title = {Can artificial systems become conscious?-Recursive embodied closure as an integrative framework.},
journal = {Frontiers in psychology},
volume = {17},
number = {},
pages = {1950309},
pmid = {42876598},
issn = {1664-1078},
abstract = {Recent advances in artificial intelligence, connectomics, and computational neuroscience increasingly raise the question of whether artificial systems-or even pure computer simulations-could become conscious. Contemporary theories emphasize embodiment, predictive processing, homeostatic regulation, and continuous interaction between cognition, body, and environment rather than isolated computation. We propose recursive embodied closure as an integrative framework linking embodied, predictive, functionalist, and dynamical approaches to consciousness. Rather than replacing existing theories, the framework identifies a shared organizational principle: the continuous recursive coupling of cognition, bodily regulation, action, and environmental feedback. We argue that consciousness may, in principle, emerge in artificial systems, including computational simulations, for three reasons. First, consciousness may depend on recursively closed loops linking cognition, body, and environment, and such loops could also be realized in robotic, neuromorphic, or virtual embodiments. Second, evidence from comparative neuroscience, convergent evolution, neurotechnology, and philosophy of mind suggests that the mechanisms relevant for consciousness are largely substrate-independent and may therefore be implemented in both biological and artificial systems. Third, although unknown physical mechanisms cannot be excluded, there is currently no empirical evidence for biological processes that are uniquely necessary for consciousness. Taken together, these considerations support the view that consciousness is best understood as an emergent property of recursively embodied, environmentally embedded cognitive systems rather than of biological matter itself. If so, biological brains, artificial neural networks, neuromorphic hardware, and even virtual agents may constitute plausible candidates for future artificial consciousness.},
}

@article {pmid42877105,
year = {2026},
author = {Park, W and Mansour, S and Kim, L and Eid, M},
title = {Decoding user satisfaction during reading of pre-defined question-answer pairs: BCI approach using delta and gamma EEG.},
journal = {Frontiers in neuroscience},
volume = {20},
number = {},
pages = {1934150},
pmid = {42877105},
issn = {1662-4548},
abstract = {INTRODUCTION: The rapid proliferation of large language models (LLMs) has intensified the need for robust evaluation frameworks to align artificial intelligence (AI) behavior with human expectations. While current reinforcement learning from human feedback (RLHF) strategies heavily rely on explicit, post-hoc self-reports, these metrics are fundamentally limited by subjective biases and user non-compliance.

METHODS: To address this, the current study presents a novel brain-computer interface (BCI) framework that systematically characterizes the continuous, implicit neural correlates of user satisfaction and dissatisfaction during a reading-based evaluative task involving pre-validated LLM-generated question-answer pairs. Sixty-four-channel electroencephalography (EEG) data were recorded from 30 participants while they read and evaluated various LLM-generated responses. To capture genuine subjective experiences, individual trials were relabeled based on post-trial 5-point Likert scale ratings into five distinct evaluative states, and spatiotemporal spectral analyses were performed. A ShallowFBCSPNet architecture was further used to evaluate single-trial classification between the very satisfied and very dissatisfied conditions under a subject-mixed cross-validation scheme.

RESULTS: Spatiotemporal spectral analyses revealed distinct neural oscillatory dynamics that temporally dissociate satisfaction and dissatisfaction. When reading highly satisfactory responses, a pronounced, sustained enhancement of low-frequency delta power was observed within the bilateral frontotemporal areas during the middle (4-8 s) and late (8-12 s) reading intervals, indexing high-order cognitive consolidation and reward processing. Conversely, processing highly dissatisfactory responses led to a localized suppression of high-frequency gamma power in the right frontotemporal region during the early interval (0-4 s), reflecting an immediate affective appraisal and expectation violation. Utilizing the ShallowFBCSPNet architecture, the system successfully discriminated between the "very satisfied" and "very dissatisfied" states, achieving an average accuracy of 81% under a subject-mixed cross-validation scheme.

DISCUSSION: These findings demonstrate that user evaluative states are encoded via distinct frequency-specific cortical networks over differing temporal windows. By bridging cognitive neuroscience and artificial intelligence, this study establishes the conceptual and empirical foundation for passive BCI-driven implicit feedback channels, laying groundwork toward future real-time paradigms for next-generation LLM alignment; the classification result should be interpreted as a feasibility demonstration rather than an estimate of real-world decoding accuracy.},
}

@article {pmid42863800,
year = {2026},
author = {Tang, E and Zhang, S and Qin, W and Xu, Y and Zheng, J and Chen, X and Wu, R and Zhao, H and Fu, Y and Zhao, C and Song, W and Li, J and Liu, C and Liu, Y and Yin, J and Huang, M and Chen, H},
title = {Attenuated attention engagement and sensorimotor resonance to social events in first-episode adolescent depression: Evidence from EEG oscillations.},
journal = {Journal of affective disorders},
volume = {417},
number = {},
pages = {122620},
doi = {10.1016/j.jad.2026.122620},
pmid = {42863800},
issn = {1573-2517},
abstract = {BACKGROUND: Depressive symptoms may disrupt social brain development during adolescence, resulting in adverse long-term quality of life. Yet, little is known regarding the cognitive and neural changes during the processing of social interactions in adolescents with first-episode depression, particularly from an allocentric perspective.

METHODS: We used electroencephalography (EEG) to assess two neural indicators during the observation of others' interactions, including sensorimotor mu suppression and occipital alpha-band activity, in 22 adolescents with first-episode major depressive disorder (MDD) and 23 healthy counterparts. Participants observed animated motions wherein actions were directed toward animate (human-like) or inanimate (rock) targets, comprising social (giving) and kinematically matched control events (disposing, taking and acquiring).

RESULTS: Healthy controls (HCs) exhibited stronger central-parietal mu suppression in the lower sub-band during observation of socially interactive giving events than disposing events. In contrast, adolescents with MDD manifested no corresponding difference in mu suppression. Additionally, depressive adolescents demonstrated significantly higher occipital alpha power when viewing social (patient) than non-social (rock) targets, and this alpha power was significantly higher than that observed in HCs. Exploratory correlation analyses revealed that attenuated sensorimotor resonance of social interactions was significantly associated with increased severity of depressive symptoms in the MDD group.

CONCLUSIONS: Adolescents with first-episode depression showed difficulties in efficiently encoding social information and implicitly extracting the social-relational meaning of observed social interactions. These findings critically extend current understanding of neurocognitive mechanisms underlying social cognitive alterations in this population, and provide potential avenues for early screening and targeted interventions of social functioning impairment in adolescence.},
}

@article {pmid42864430,
year = {2026},
author = {Chen, F and Wang, Q and Wang, L and Bhatia, S},
title = {Simulated responses from large language models are weak proxies for human economic preferences across 74 countries.},
journal = {Nature human behaviour},
volume = {},
number = {},
pages = {},
pmid = {42864430},
issn = {2397-3374},
support = {72631009, 72322009, 72173113//National Natural Science Foundation of China (National Science Foundation of China)/ ; 72371226//National Natural Science Foundation of China (National Science Foundation of China)/ ; Grant SES-1847794//National Science Foundation (NSF)/ ; },
abstract = {Large-scale surveys are central to investigating economic preferences but are costly and slow. Can large language model (LLM)-simulated responses substitute for human survey data? We conducted a virtual Global Preference Survey, simulating six experimentally validated economic preferences for 67,201 matched human respondents across 74 countries. Across 20 LLM configurations with diverse prompts and robustness checks, we found systematic but modest individual-level alignment (87.5% of Spearman correlations have 95% confidence intervals (CIs) above zero), with LLMs recovering the direction of associations in 77.8% (95% CI, (71.6%, 83.1%); CI-based consistency ratio) of gender, age and cognitive-ability dimensions, but only 48.1% (95% CI, (45.1%, 51.2%)) of cultural dimensions. Ablation and error decompositions suggest weaker alignment when removing demographic information than when removing country labels, and in many models alignment is weaker in economically undeveloped and less digitally connected countries. LLM-simulated responses may recover some individual- and group-level patterns but are not yet reliable proxies for cross-national representative surveys in macro-level inference or policy evaluation.},
}

@article {pmid42865765,
year = {2026},
author = {Lee, AH and Lee, J and Leung, V and Larson, L and Nurmikko, A},
title = {Distributed implantable microchips as wireless networks for neural sensing and stimulation.},
journal = {Frontiers in neuroscience},
volume = {20},
number = {},
pages = {1855768},
pmid = {42865765},
issn = {1662-4548},
abstract = {We examine ongoing research for implantable sensors and stimulators for future brain-computer applications and give a review on the concept where grain-of-sand size silicon microchips form the individual nodes in a multichannel wireless implant. In such an approach, each chip can perform the function of a stand-alone independent sensor if equipped with active microelectronic circuits designed to detect or stimulate target populations of neurons and capability for wireless communication. Delivering power to individual as well as spatially distributed ensembles of microchips requires a method for efficient transcutaneous/transcranial wireless energy transfer approach. Here we offer examples from authors' experience and comparisons to contemporary approaches as well as the challenges ahead in developing clinically viable wireless cortical chip-in-the-brain interfaces.},
}

@article {pmid42866011,
year = {2026},
author = {Trocellier, D and Kojima, S and N'kaoua, B and Lotte, F},
title = {Visual cues in MI-BCI induce biases in EEG classification models with deep learning but not with standard machine learning.},
journal = {Frontiers in neuroergonomics},
volume = {7},
number = {},
pages = {1804143},
pmid = {42866011},
issn = {2673-6195},
abstract = {OBJECTIVE: Deep learning classification methods are increasingly popular in the brain-computer interface (BCI) field due to their ability to learn complex, non-linear representations from EEG data. Unlike traditional machine learning approaches, which typically rely on hand-crafted features, deep learning models autonomously extract spatio-temporal features directly from the raw or minimally preprocessed signal. However, this flexibility may come at the cost of reduced interpretability and increased susceptibility to confounding factors. This study aims to investigate whether the inclusion of task-related visual cues in the input EEG window biases deep learning models by leading them to exploit irrelevant features, thereby compromising their neurophysiological validity.

APPROACH: This study investigates how including EEG signals measured during visual cue presentation on classification accuracy and feature extraction in motor imagery (MI)-based BCIs. Three classification pipelines were evaluated: CSP+LDA, a traditional machine learning approach; DeepConvNet, a convolutional neural network for EEG classification; and REVE, a Transformer-based EEG foundation model. All methods are trained and tested on two large scale open-access EEG datasets using two epoching strategies, with or without including the visual cue onset, to assess their impact on model performance and neurophysiological relevance. Attribution maps (e.g., Saliency maps) and CSP patterns are used to interpret the features learned by each model.

MAIN RESULTS: Including the visual cue onset significantly improved the classification accuracy of DeepConvNet and REVE. However, attribution analysis showed that the model mainly relied on early parietal activations, most likely cue-related visual evoked potentials, rather than MI-related sensorimotor rhythms (ERD/ERS), which were more prominent when the cue onset was excluded. In contrast, CSP+LDA showed minimal accuracy changes across conditions, with consistent motor cortex activation patterns.

SIGNIFICANCE: These findings demonstrate that higher classification performance does not necessarily imply neurophysiological validity. In particular, deep learning models may exploit confounding signals, such as visual cue processing, which undermines their interpretability and applicability in asynchronous or real-world BCIs, in which there is no such cue. We advocate for systematically applying explainability tools in BCI research to ensure classifiers rely on task-relevant brain activity. Additionally, we emphasize the importance of excluding visual cue periods from EEG segments containing reposenses evoked by the visual cue onset from the input to deep learning models.},
}

@article {pmid42866037,
year = {2026},
author = {Zhou, C and Ding, X and Zhao, J and Gu, X and Xu, Z},
title = {Terahertz liquid networks: a mesoscopic passive architecture for circumventing physical deadlocks in brain-computer interfaces.},
journal = {Nanoscale advances},
volume = {},
number = {},
pages = {},
pmid = {42866037},
issn = {2516-0230},
abstract = {Brain-computer interfaces (BCIs) for major neurological disorders face physical and physiological limitations. Mechanical mismatch in rigid microelectrode arrays precipitates chronic immune rejection and irreversible central nervous system damage, whereas bandwidth scaling in traditional wireless links inevitably breaches the 1 °C tissue thermal safety threshold. To circumvent this deadlock among bandwidth, heat dissipation, and biocompatibility, this review proposes the "terahertz liquid network" architecture. Utilizing injectable plasmonic nanoparticles as free-floating relay nodes, this architecture fundamentally abolishes the mechanical shear stresses characteristic of conventional implant interfaces. Exploiting broadband and dielectric properties of terahertz (THz) waves, the network achieves cross-scale bioelectronic integration across both physical and physiological boundaries. For communication and decoding, the system bypasses the strong THz absorption barrier of brain tissue via non-line-of-sight multipath scattering and THz-photoacoustic cross-modal transduction. Integrating these mechanisms with passive backscattering fulfills the terabits-per-second (Tbps) absolute bandwidth requirements dictated by full-cortical, high-fidelity motor decoding. For neuromodulation, specific THz photon energies are directed at voltage-gated ion channels to induce targeted non-thermal resonance, mediating precise, subcellular-level interventions devoid of macroscopic current overflow. Through an analysis of typical clinical scenarios-namely, audiovisual perception reconstruction and physical neural injury rehabilitation-this review demonstrates the feasibility of this architecture in resolving the conflict between high-throughput concurrency and stringent physiological constraints. Ultimately, the terahertz liquid network reconfigures the neural interface paradigm from a macroscopic active solid-state to a mesoscopic passive liquid-state. This transition grants the system superior biocompatibility and endows it with disease-modifying potential targeting underlying pathologies, establishing a trajectory for next-generation BCIs.},
}

@article {pmid42866729,
year = {2026},
author = {Hassan, ST and Faheem, MSB and Yaqubi, AJ and Brhaish, MS},
title = {Brain-computer interfaces and neuroprostheses for severe speech impairment in amyotrophic lateral sclerosis.},
journal = {Annals of medicine and surgery (2012)},
volume = {88},
number = {10},
pages = {6221-6224},
pmid = {42866729},
issn = {2049-0801},
abstract = {Amyotrophic lateral sclerosis (ALS) is a relentlessly progressive neurodegenerative disorder that frequently culminates in severe dysarthria and anarthria, profoundly isolating patients and diminishing their quality of life. Traditional speech therapy and conventional augmentative and alternative communication devices rely on residual motor control and become ineffective in late-stage ALS. In response, brain-computer interfaces and speech neuroprostheses have emerged as transformative solutions. By decoding neural signals directly from the motor cortex and translating them into synthesized speech, these devices bypass compromised neuromuscular pathways. Recent high-impact clinical trials have demonstrated unprecedented success, achieving high-speed decoding and integrating artificial intelligence-driven voice banking to preserve a patient's unique vocal identity. However, significant translational challenges remain, including signal degradation, surgical risks, high socioeconomic costs, and ethical dilemmas regarding patient autonomy and data privacy. This perspective reviews the current state-of-the-art in speech neuroprosthetics, analyzes the barriers to widespread clinical implementation, and advocates for increased interdisciplinary collaboration and funding to establish these technologies as a standard of care for patients with severe speech impairments.},
}

@article {pmid42869466,
year = {2026},
author = {Hamard, R and Hug, F and Dick, TJM},
title = {Passive ankle exoskeleton assistance selectively modulates triceps surae neural control during balance.},
journal = {Journal of the Royal Society, Interface},
volume = {23},
number = {243},
pages = {},
doi = {10.1098/rsif.2025.1305},
pmid = {42869466},
issn = {1742-5662},
support = {//ISB International Travel Grant/ ; //The Company of Biologists Travelling Fellowship/ ; },
mesh = {Humans ; *Muscle, Skeletal/physiology/innervation ; *Exoskeleton Device ; *Postural Balance/physiology ; Male ; Female ; Adult ; *Ankle/physiology ; Electromyography ; Biomechanical Phenomena ; },
abstract = {Wearable assistive devices have potential to improve human balance, but their effectiveness remains limited by an incomplete understanding of how mechanical assistance influences neuromuscular control. This study aimed to determine the neural adaptations in the control of lower leg muscles in response to passive ankle exoskeleton assistance during standing balance tasks with varying levels of difficulty. Lower limb kinematics and kinetics were recorded to assess changes in joint dynamics and balance performance. Neural drive to the muscles was assessed by decomposing high-density electromyography signals into motor unit spike trains. Ankle exoskeleton assistance, provided by a simple spring-loaded passive device, did not alter or improve balance performance and had minimal effects on lower limb kinematics. However, exoskeleton assistance affected triceps surae neural control, but only in the soleus, by reducing neural drive, while the gastrocnemii showed no detectable changes in neural control. Together, these results demonstrate that passive exoskeleton assistance alters neural control of the triceps surae in a muscle-specific manner during balance. Understanding muscle-specific adaptations can support more effective exoskeleton design and the development of their control strategies informed by myoelectrical signals.},
}

Humans
*Muscle, Skeletal/physiology/innervation
*Exoskeleton Device
*Postural Balance/physiology
Male
Female
Adult
*Ankle/physiology
Electromyography
Biomechanical Phenomena

@article {pmid42861763,
year = {2026},
author = {Palthe, S and van Merode, NAM and Kolenburg, J and Marcelissen, T and Steffens, MG and van Koeveringe, GA and Heesakkers, JP and Witte, LPW},
title = {Bladder Contractility Classification Before Bladder Outlet Surgery: Do BCI, Wmax and the Maastricht-Hannover Nomogram Agree and Does It Matter for Outcome?.},
journal = {Neurourology and urodynamics},
volume = {},
number = {},
pages = {},
doi = {10.1002/nau.70489},
pmid = {42861763},
issn = {1520-6777},
abstract = {OBJECTIVES: The aim of this study was to assess the concordance between three classification systems (bladder contractility index, Maximum Watt's factor and Maastricht-Hannover nomogram) in allocating men to bladder contractility subgroups. Secondary aim was to explore whether preoperative contractility classification was associated with short-term flowmetry outcomes after BOP.

MATERIALS AND METHODS: This multicenter retrospective cohort study included men who underwent BOP (2010-2016) for lower urinary tract symptoms with available baseline urodynamic study and 6-week postoperative flowmetry. Patients were classified according to BCI, Wmax and MH-nomogram into very weak (BCI < 50, Wmax < 5, MH-nomogram < 10th percentile), weak (BCI: 50-100, Wmax: 5-10, MH-nomogram 10-25th percentile) and normal detrusor contractility (BCI > 100, Wmax > 10, MH-nomogram > 25th percentile) subgroups. Concordance was quantified using percentage agreement and Cohen's linear weighted kappa (κ). Postoperative flowmetry outcomes were compared between subgroups within each classification and pre-postoperative differences were determined for each subgroup.

RESULTS: One hundred and thirty three patients were included (median age 69 years). Agreement was highest between BCI and Wmax (78.9%, κ = 0.61, substantial), while agreement was weak between Wmax and MH-nomogram (48.9%, κ = 0.31) and negligible between BCI and MH-nomogram (37.6%, κ = 0.05). All subgroups showed improvement Qmax and PVR parameters after BOP, with no significant association found between preoperative contractility grade and postoperative outcomes.

CONCLUSIONS: BCI, Wmax and the MH-nomogram show only limited concordance, particularly involving the MH-nomogram, and cannot be regarded as interchangeable. Contractility classification was not associated with postoperative outcomes, and all subgroups improved after BOP. These findings underscore the need for a clinically relevant urodynamic consensus definition of DU and support shared decision-making over classification-based patient selection alone.},
}

@article {pmid42863207,
year = {2026},
author = {Ma, Y and Li, Y and Zhang, S and Liang, L and Li, G},
title = {Emerging promise in artificial vision research: Unveiling transformative potential through brain-computer interfaces.},
journal = {Neural networks : the official journal of the International Neural Network Society},
volume = {206},
number = {Pt A},
pages = {109693},
doi = {10.1016/j.neunet.2026.109693},
pmid = {42863207},
issn = {1879-2782},
abstract = {Brain-computer interface (BCI) technology has driven revolutionary advancements in the field of artificial vision, presenting new opportunities for individuals with visual impairments. This review summarizes the key technological developments, clinical applications, and future challenges of BCI in artificial vision. In the domain of signal acquisition, both invasive techniques (such as MEA, ECoG, and DBS) and non-invasive techniques (including EEG, MEG, fMRI, and fNIRS) have significantly improved signal resolution and stability. Advances in visual encoding have facilitated the more efficient transmission of visual information. Input BCIs, including technologies based on SSVEP and P300, as well as output BCIs involving motor and speech decoding, are progressing rapidly. Clinically, third-generation retinal prostheses (such as Argus II and PRIMA) have helped patients with retinal degenerative diseases regain partial light perception. Based on closed-loop neural feedback, it is expected to help patients with cortical blindness achieve the remodeling of the optic nerve. Nevertheless, technical challenges remain, particularly regarding the long-term stability of neural signals and biocompatibility issues. Emerging solutions such as anti-scarring nanocoatings, self-powered wireless systems, neuromorphic spike encoding, and multimodal sensory integration are addressing several of these challenges. Looking ahead, advancements in neuro-augmented reality are expected to further propel progress in the field. The future of BCI development will depends on continued interdisciplinary collaboration and global cooperation to drive innovation and provide more efficient, accessible visual restoration solutions for patients.},
}

@article {pmid42863253,
year = {2026},
author = {Wang, C and Ding, J and Qi, X and Jialin, A and Xue, C and Hu, X and Shao, Y and Cheng, Y and Deng, W},
title = {Medial prefrontal cortex circuitry in inflammation-associated major depressive disorder: Brain-immune interactions and translational rationale for noninvasive neuromodulation.},
journal = {Asian journal of psychiatry},
volume = {125},
number = {},
pages = {105181},
doi = {10.1016/j.ajp.2026.105181},
pmid = {42863253},
issn = {1876-2026},
abstract = {Major depressive disorder (MDD) is a leading cause of disability worldwide, characterized by marked clinical and biological heterogeneity. Inflammation represents one key mechanism contributing to MDD, with a substantial proportion of patients exhibiting elevated inflammatory biomarkers. Inflammatory signaling affects neurotransmitter systems and neural circuits involved in reward, motivation, arousal, and threat processing, thereby linking to neurovegetative and motivational symptoms and reduced treatment responsiveness. The medial prefrontal cortex (mPFC) is a recurring node across inflammation-related symptom dimensions and provides a tractable focus for examining how inflammatory burden intersects with prefrontal, striatal, and limbic dysfunction. Sex may further moderate this brain-immune relationship through differences in stress susceptibility, brain structure and function, and inflammatory activity. Given the inconsistent effects of anti-inflammatory agents, noninvasive neuromodulation offers a targeted strategy for inflammation-associated MDD. Bottom-up transcutaneous auricular vagus nerve stimulation may regulate autonomic-immune signaling and ascending network pathways, whereas top-down high-definition transcranial direct current stimulation may modulate prefrontal plasticity and engage mPFC-centered circuitry. In this review, we synthesize evidence on peripheral and central inflammatory involvement in MDD, with emphasis on mPFC-centered circuit vulnerability and sex as a moderator of brain-immune interactions, and evaluate the translational rationale for complementary bottom-up and top-down stimulation in patients with elevated inflammatory burden. We also highlight key priorities for future research, including patient enrichment, outcome alignment, target engagement, parameter optimization, and sex-related variables to advance precision and stratified therapeutics in depression.},
}

@article {pmid42863491,
year = {2026},
author = {Chen, B and Deng, C and Ge, H and Su, X and Shen, Y},
title = {MTDANet: lightweight multi-scale temporal hierarchy and distance-regularized linear attention network for motor imagery brain-computer interfaces.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/aeb330},
pmid = {42863491},
issn = {1741-2552},
abstract = {OBJECTIVE: Motor-imagery (MI) electroencephalography (EEG) offers a non-invasive route to brain-computer interaction, yet reliable single-trial decoding remains difficult because task-related patterns are weak, nonstationary, and distributed over multiple temporal scales. This study investigates whether a staged local-to-global model can integrate fine temporal-spatial structure with longer-range context without relying on unrestricted self-attention.

APPROACH: We propose MTDANet, an architecture that combines multi-scale temporal-spatial feature extraction, factorized temporal consolidation, and distance-regularized recurrent attention. The model was evaluated on three public MI datasets under subject-specific and leave-one-subject-out (LOSO) protocols. Separately initialized models were also trained on P300 and steady-state visual evoked potential (SSVEP) datasets to examine broader architectural applicability. Participant-level paired analyses used two-sided paired t tests with within-setting Holm correction.

MAIN RESULTS: MTDANet achieved subject-specific accuracies of 85.61 ± 8.32%, 90.85 ± 7.62%, and 87.79 ± 13.93% on BCI Competition IV 2a, BCI Competition IV 2b, and BNCI2014_002, respectively. Relative to the highest-mean baseline, the paired improvements were supported after within-setting Holm correction on BCI Competition IV 2a (3.56 percentage points, adjusted p = 0.0097) and 2b (1.62 points, adjusted p = 0.0168). Under LOSO evaluation, the corresponding accuracies were 63.31 ± 15.74%, 79.77 ± 7.57%, and 79.96 ± 11.95%. The 2.08-point BNCI2014_002 difference had a 95% confidence interval of 0.42-3.73 points and an adjusted p value of 0.0387. These three comparisons did not remain significant in a broader 56-comparison Holm sensitivity analysis. The P300 and SSVEP models attained the highest observed mean accuracies.

SIGNIFICANCE: The results provide within-setting evidence for staged local-to-global temporal modeling in offline MI-EEG evaluations.},
}

@article {pmid42863775,
year = {2026},
author = {Lin, YG and Ni, J and Yu, H and Li, T},
title = {Increased Morning Cortisol Levels across Different Mood States in Bipolar Disorder: A Systematic Review and Meta-Analysis.},
journal = {Neuroscience and biobehavioral reviews},
volume = {},
number = {},
pages = {107027},
doi = {10.1016/j.neubiorev.2026.107027},
pmid = {42863775},
issn = {1873-7528},
abstract = {OBJECTIVE: Bipolar disorder (BD) is associated with hypothalamic-pituitary-adrenal (HPA) axis dysregulation. However, findings regarding morning cortisol levels across different mood states remain inconsistent. This systematic review and meta-analysis aimed to synthesize evidence on morning cortisol levels in BD across different mood states, while considering the effects of various clinical moderators.

METHODS: A systematic search of seven databases was conducted for articles published up to August 31, 2026. Studies were included if they reported cortisol levels measured under eligible morning sampling protocols in BD patients and healthy controls. A total of 39 studies (n = 3,611) were included. A multilevel mixed-effects meta-analysis was applied to account for the dependence of mood-state subgroups within studies. Moderator analyses were conducted via meta-regression.

RESULTS: Compared with healthy controls, BD patients exhibited significantly elevated morning cortisol levels, both overall and across specific mood states. In exploratory moderator analyses, earlier age at onset was associated with higher morning-cortisol, whereas a greater proportion of participants receiving medication was associated with lower morning cortisol after adjustment for mood state.

CONCLUSION: BD is associated with a significant elevation in morning cortisol across mood states. This pattern is consistent with a potential trait-related component of HPA-axis dysregulation, although its expression may vary with clinical state and sampling conditions.},
}

@article {pmid42855851,
year = {2026},
author = {Heyat, MBB and Sultana, A and Rehman, AU and Begum, A and Sawan, M},
title = {Ensemble-Based Machine Learning Models for Classifying Intervention-Related Psychological Outcome Patterns in Menopausal Women Treated With Pimpinella anisum.},
journal = {Brain and behavior},
volume = {16},
number = {10},
pages = {e71787},
pmid = {42855851},
issn = {2162-3279},
support = {10318A992001//Westlake University/ ; //Hangzhou Postdoctoral Research/ ; 10318A992001//Westlake University/ ; 10318A992001//Westlake University/ ; },
mesh = {Humans ; Female ; *Menopause/psychology ; *Machine Learning ; Middle Aged ; Random Forest ; Boosting Machine Learning Algorithms ; Predictive Learning Models ; Classification Algorithms ; Depression ; Anxiety ; Prediction Algorithms ; },
abstract = {BACKGROUND: Menopause, a natural biological transition in the life of women, is often associated with significant psychological disturbances such as depression, anxiety, and stress. However, early identification and management of psychological disorders in menopausal women remain under-addressed due to limited specialist access and a lack of systematic screening approaches.

PURPOSE: This study aimed to develop ensemble-based machine learning (ML) models to classify intervention-related clinical and psychological outcome patterns in menopausal women using clinical and questionnaire-based data collected during a controlled intervention study.

METHOD: Psychological and menopausal symptoms were assessed using standardized tools, including the Depression Anxiety Stress Scale-21 (DASS-21), modified Kupperman index (MKI), vaginal health index (VHI), and menopause symptoms treatment satisfaction questionnaire (MS-TSQ). Statistical analyzes were performed using chi-square tests and independent t-tests to evaluate group differences. Three ensemble classifiers, random forest (RF), gradient boosting (GB), and AdaBoost (AB), were evaluated using repeated cross-validation.

RESULT: Among these, the RF classifier achieved the best performance with a sensitivity of 93.3%, a specificity of 90.0%, an accuracy of 91.7%, a precision of 90.3%, and an F1-score of 91.8%.

CONCLUSION: These results indicate that ensemble ML models, particularly RF, can serve as effective supplementary tools for psychological disorder risk stratification in menopausal care, aiding clinicians in decision-making. However, these models are intended to complement, not replace, clinical assessments.},
}

Humans
Female
*Menopause/psychology
*Machine Learning
Middle Aged
Random Forest
Boosting Machine Learning Algorithms
Predictive Learning Models
Classification Algorithms
Depression
Anxiety
Prediction Algorithms

@article {pmid42856243,
year = {2026},
author = {Conlan, EC and Foli, C and Memberg, WD and Herring, EZ and Sweet, JA and Ajiboye, AB},
title = {Assessing the Relative Impact of Grasp and Object on Inferior Frontal Gyrus Activity during a Grasping Task.},
journal = {bioRxiv : the preprint server for biology},
volume = {},
number = {},
pages = {},
doi = {10.64898/2026.06.30.735663},
pmid = {42856243},
issn = {2692-8205},
abstract = {The lateral grasp network, responsible for translating visual properties of an object to execution of a motor act, is comprised of the anterior intraparietal area (AIP), area F5, and the primary motor cortex (M1). Non-human primate studies of F5 have shown that it encodes a wide range of hand positions and object properties. Human studies in F5's human homologue, the inferior frontal gyrus (IFG), have leveraged this area's ability to encode grasp-object pairs for the purposes of Brain Machine Interface (BMI) control. However, whether modulation is driven by grasp, object, or the interaction between grasp and object is unclear. In the present study, intracortical IFG signals were recorded from human participants with tetraplegia during a motor visualization task where grasp and object were varied. Grasp was found to be the predominant factor driving modulation of IFG signals. Object was found to only be weakly represented in neural data. However, object contribution peaked earlier than grasp contribution, indicating early integration of object information. Grasp-object interactions were also found to have a significant impact. Cortical separation between object conditions varied based on the grasp used. In addition, subspace analysis showed that the underlying neural population structure associated with each grasp type was significantly different from one another. Despite the impact of grasp-object interactions, the present study demonstrates that grasp is the primary driver of inferior frontal gyrus neural modulation.},
}

@article {pmid42856741,
year = {2026},
author = {Settgast, T and Shriki, O and Northoff, G and Zilio, F and Borgheai, SB and Shahriari, Y and Kübler, A},
title = {All Roads Lead to Rome: Resting-State Criticality as a Neurophysiological Marker of Brain-Computer Interface (BCI) Performance in People with Amyotrophic Lateral Sclerosis (ALS) and Healthy Participants.},
journal = {Research square},
volume = {},
number = {},
pages = {},
doi = {10.21203/rs.3.rs-10422321/v1},
pmid = {42856741},
issn = {2693-5015},
abstract = {This exploratory study investigates the relationship between resting-state brain dynamics (EEG) and subsequent brain-computer interface (BCI) performance in people with amyotrophic lateral sclerosis (ALS) and healthy participants. Grounded in the critical brain hypothesis, we examined whether neurophysiological markers of criticality in resting-state EEG correlate with subsequent BCI performance. We analyzed neuronal avalanche dynamics, multiscale criticality measures, the spectral slope, and Lempel-Ziv complexity across two datasets: a longitudinal single-case study of a person in the locked-in state after a disease history with ALS (Dataset 1) and longitudinal data from 5 people with ALS and single sessions from 6 healthy participants (Dataset 2). Neuronal avalanche dynamics, the spectral slope, and Lempel-Ziv complexity consistently correlated with subsequent BCI performance in people with ALS. Conversely, healthy participants showed inverted correlation patterns for the neuronal avalanche dynamics. While group comparisons were not statistically significant, effect size analyses indicated supercritical brain dynamics in people with ALS. We integrate these findings into the B-ARC (BCI-Aptitude through Resting-State Criticality) framework proposing an inverted U-shaped relationship between the proximity of resting-state brain activity to criticality and BCI control. In this framework, healthy participants approach the critical optimum from a subcritical regime, while people with ALS appear to approach it from a putatively supercritical regime characterized by cortical hyperexcitability. These findings and the proposed framework provide potential resting-state predictors of BCI aptitude in people with ALS and may inform the detection of so-called Windows of Consciousness .},
}

@article {pmid42858309,
year = {2026},
author = {Li, Y and Feng, J and Hu, W and Zeng, Z and Pan, Q and Wang, J and Li, M and Elbert, T and Karl, A and Zang, Y},
title = {Dismantling Psychological Interventions for Youth with Complex Trauma: A Systematic Review and Component Network Meta-Analysis.},
journal = {Psychotherapy and psychosomatics},
volume = {},
number = {},
pages = {1},
doi = {10.1159/pps/aemag009},
pmid = {42858309},
issn = {1423-0348},
abstract = {INTRODUCTION: Complex trauma-related posttraumatic stress disorder (PTSD) and comorbidities pose a significant mental health burden in children. However, the comparative effectiveness of psychological interventions and shared therapeutic components remains unclear.

METHODS: We conducted a systematic review, network, and component network meta-analysis of randomized controlled trials of psychological interventions for children and adolescents with complex trauma. PubMed, Embase, Web of Science, PsycINFO, and Cochrane Library were searched up to September 2024. PTSD was the primary outcome; retention and comorbid symptoms (depression, anxiety, dissociation, externalizing problems) were secondary outcomes.

RESULTS: Forty-eight studies (6,017 participants) were included in this review. Multimodal group interventions (SMD -1.68, 95% confidence interval [CI] -3.04 to -0.32), phase-based trauma-focused cognitive behavioral therapy (TF-CBT) with parental involvement (-1.12, -1.53 to -0.70), TF-CBT (-1.06, -1.59 to -0.52), non-TF-CBT with parental involvement (-0.93, -1.70 to -0.17), eye movement desensitization and reprocessing (-0.85, -1.60 to -0.10), and phase-based TF-CBT (-0.79, -1.20 to -0.39) significantly reduced PTSD. Exposure emerged as the only significantly effective component for PTSD (-0.85, 95% credible interval [CrI] -1.60 to -0.06). Efficient component combinations included affective modulation with a therapeutic foundation of psychoeducation and nonspecific elements for PTSD (-0.97, 95% CrI -1.92 to -0.01) and exposure plus parental involvement with the same foundation for depression (-2.11, 95% CrI -4.06 to -0.20).

CONCLUSION: Trauma-focused approaches were the most effective treatments for pediatric PTSD and emotional comorbidities related to complex trauma. A more cost-effective intervention approach may prioritize exposure techniques, incorporate affective modulation, and include parental involvement when feasible.},
}

@article {pmid42858322,
year = {2026},
author = {Yang, Y and Zhang, X and Zan, X and Niu, C and Wei, X and Feng, P and Li, G and Gao, M},
title = {A Clearance-Repair-Activation Framework for Alzheimer's Disease: A Perspective and Validation Roadmap.},
journal = {Aging and disease},
volume = {},
number = {},
pages = {},
doi = {10.14336/AD.2026.1040},
pmid = {42858322},
issn = {2152-5250},
abstract = {Alzheimer's disease (AD) is a heterogeneous clinicobiological syndrome involving impaired protein clearance, neuroinflammation, synaptic and cellular injury, network dysfunction, and progressive functional decline. Amyloid-directed antibodies provide an evidence-based treatment backbone for selected patients with early symptomatic AD, but adjunctive strategies may be needed to address other disease mechanisms. This Perspective proposes a clearance-repair-activation framework for hypothesis-driven research. Clearance includes established pharmacological treatment and risk management as well as investigational drainage approaches; lymphaticovenous anastomosis (LVA) and deep cervical lymphaticovenous anastomosis (dcLVA) are considered only as research options within this broader domain. Repair includes neural stem-cell (NSC), induced pluripotent stem-cell-derived, engineered-cell, and secretome-based strategies, for which AD evidence remains predominantly preclinical. Activation encompasses invasive and non-invasive stimulation, sensory entrainment, neurofeedback, brain-computer interface (BCI) monitoring, and cognitive rehabilitation; BCI is positioned mainly as an experimental monitoring and adaptive rehabilitation platform. We define falsifiable links between clearance and repair readiness and between repair markers and activation responsiveness, consider adaptive and parallel alternatives to fixed sequencing, and integrate disease stage, heterogeneity, regulation, and combination-trial design. Informatics is limited to trial infrastructure and safety governance. The validation roadmap specifies success criteria, stop rules, interaction testing, clinically meaningful outcomes, and long-term surveillance.},
}

@article {pmid42860745,
year = {2026},
author = {Lee, C and Park, S and Lee, C and Lee, D and Ha, H and Shin, H and Kim, T and Cho, C and Byun, K and Lee, S},
title = {Associations of adiposity and physical fitness with hippocampaldependent mnemonic discrimination in postmenopausal women.},
journal = {Physical activity and nutrition},
volume = {30},
number = {3},
pages = {162-172},
pmid = {42860745},
issn = {2733-7545},
support = {//Ministry of Education/ ; 2023S1A5A2A21084054//National Research Foundation of Korea/ ; },
abstract = {PURPOSE: This study examined the associations of adiposity indices and physical fitness components with domain-specific cognitive function in postmenopausal women, focusing on hippocampus-dependent mnemonic discrimination.

METHODS: The study involved 68 postmenopausal women aged ≥ 40 years who were at least one year postmenopausal. Body composition was assessed using dual-energy X-ray absorptiometry, and adiposity indices comprised body fat mass, body fat percentage, fat mass index, fat-to-lean mass ratio, and android-to-gynoid fat ratio. Cognitive function was assessed using the color-word Stroop task and the object and spatial versions of the Mnemonic Discrimination Task. Physical fitness was evaluated based on grip strength, sit-and-reach flexibility, and peak oxygen uptake. Correlation and hierarchical multiple linear regression analyses were performed (n = 61).

RESULTS: Adiposity indices were not significantly associated with Stroop performance. Body fat mass and fat mass index were negatively associated with object-version recognition score and spatial-version lure discrimination index, and body fat mass was also negatively associated with spatial-version recognition score. Spatial-version lure discrimination index was positively correlated with flexibility and peak oxygen uptake. In hierarchical regression, adding fat mass index and peak oxygen uptake to age significantly improved the model predicting spatial-version lure discrimination index; peak oxygen uptake was significant at this step. However, flexibility provided no significant additional explanatory power, and no predictor remained significant in the final model.

CONCLUSION: Hippocampus-dependent mnemonic discrimination in postmenopausal women may be related to adiposity and overall fitness status at the level of simple associations; however, the independent contribution of specific fitness components requires further investigation.},
}

@article {pmid42852247,
year = {2026},
author = {Guan, Y and Gao, Y},
title = {Neurorehabilitation for cognitive and motor recovery: mechanisms, interventions, and translational advances.},
journal = {Frontiers in neurology},
volume = {17},
number = {},
pages = {1828447},
pmid = {42852247},
issn = {1664-2295},
mesh = {Humans ; *Recovery of Function/physiology ; *Neurological Rehabilitation/methods/trends ; Neuronal Plasticity/physiology ; Translational Research, Biomedical ; *Stroke Rehabilitation/methods ; Cognitive Training ; Stroke/physiopathology/complications ; Animals ; },
abstract = {This narrative review synthesizes recent advances in neurorehabilitation, focusing on mechanisms of neural plasticity, glial cell function, and interventions for cognitive and motor recovery. We explore multidisciplinary approaches, including cognitive training, motor rehabilitation, neurostimulation, and regenerative therapies, highlighting their translational potential. Special attention is given to clinically relevant rehabilitation modifiers and complications, including spasticity, frailty, and post-stroke epilepsy, alongside strategies for enhancing quality of life and social reintegration. Emerging technologies such as brain-computer interfaces, virtual reality, and robotics are discussed for their role in personalized rehabilitation. This review emphasizes evidence-based, patient-centered interventions to enhance functional recovery and quality of life, identifying future directions for personalized neurorehabilitation across stroke survivors, elderly, and frail populations.},
}

Humans
*Recovery of Function/physiology
*Neurological Rehabilitation/methods/trends
Neuronal Plasticity/physiology
Translational Research, Biomedical
*Stroke Rehabilitation/methods
Cognitive Training
Stroke/physiopathology/complications
Animals

@article {pmid42852356,
year = {2026},
author = {Song, W and Si, S and Wang, Y and Chang, H and Tang, X},
title = {Application of Artificial Intelligence in the Prediction and Management of Stroke Rehabilitation.},
journal = {Cyborg and bionic systems (Washington, D.C.)},
volume = {7},
number = {},
pages = {0702},
pmid = {42852356},
issn = {2692-7632},
abstract = {Artificial intelligence is being studied across the stroke care pathway, but evidence from acute diagnosis, early prognosis, rehabilitation-outcome prediction, and long-term management is often discussed without clearly separating these clinical tasks. This structured narrative review adds an integrated framework that maps each application to its decision point, relevance to rehabilitation, validation level, and readiness for clinical use. Unlike previous technology-centered reviews, it explicitly separates acute prognostic evidence from rehabilitation-specific evidence and distinguishes technical performance from transportability, clinical impact, and rehabilitation benefit. Acute imaging and prognostic models may provide baseline information for later rehabilitation planning, but their diagnostic or prognostic performance does not establish rehabilitation efficacy. Most rehabilitation models and robotic, virtual-reality, brain-computer interface, wearable, and home-monitoring applications remain supported mainly by internal validation or early exploratory studies. Small or selected datasets, limited external and prospective validation, uncertain workflow effects, and sparse patient-centered outcomes continue to constrain clinical interpretation.},
}

@article {pmid42854016,
year = {2026},
author = {Zhu, SY and Xiao, QX and Duan, ZY and Ma, YY and Zhao, PC and Chen, HX and Bao, YJ and Lei, JM and Wang, L and Zhou, H and Ma, RX and Pu, HL and Wu, KF and Xiong, LL and Du, RL},
title = {Bionic sleep therapy combined with multimodal management for chronic insomnia with alcohol withdrawal-related symptoms: A case report.},
journal = {Medicine},
volume = {105},
number = {41},
pages = {e50991},
doi = {10.1097/MD.0000000000050991},
pmid = {42854016},
issn = {1536-5964},
support = {No. GZWJWPF2025021//Guizhou Provincial Health Commission Clinical Key Specialty Construction “Peak Climbing Plan†Project/ ; Not applicable//Young Elite Scientists Sponsorship Program by the Guizhou Association for Science and Technology/ ; QNB(2026)001//Youth Science Fund Project (Category B) of the Basic Research Program of the Science and Technology Plan Project of Guizhou Province/ ; },
mesh = {Humans ; Male ; Aged ; *Sleep Initiation and Maintenance Disorders/therapy ; *Substance Withdrawal Syndrome/therapy ; Hypnotics and Sedatives/administration & dosage/therapeutic use ; Combined Modality Therapy ; Dexmedetomidine/administration & dosage/therapeutic use ; Propofol/administration & dosage/therapeutic use ; Cognitive Behavioral Therapy/methods ; *Alcoholism/complications ; },
abstract = {RATIONALE: Patients who use alcohol as a sleep aid may experience withdrawal-related insomnia and hyperarousal, which can reinforce continued drinking. Bionic sleep therapy (BST) is an exploratory intervention using intravenous anesthetic and sedative agents under close monitoring to induce a brief, controllable, sleep-like state. It may provide short-term support for sleep initiation during early alcohol withdrawal.

PATIENT CONCERNS: A 71-year-old man had chronic insomnia for over 15 years and consumed 100 to 200 mL of Chinese baijiu nightly to initiate sleep, with additional alcohol after nocturnal awakenings. Repeated attempts to stop drinking were followed by worsening insomnia, restlessness, palpitations, diaphoresis, and intense alcohol craving, which improved after drinking.

DIAGNOSES: Chronic insomnia with recurrent mild alcohol withdrawal-related symptoms and a high risk of complicated alcohol withdrawal.

INTERVENTIONS: The patient underwent 2 monitored BST sessions on hospital days 1 and 2 under continuous cardiorespiratory monitoring. Propofol (1 mg/kg) was administered intravenously for rapid induction, followed by a 60-minute dexmedetomidine infusion via an IV pump at 0.2 to 0.4 μg/kg/h. Multimodal treatment was intensified on hospital day 2, including cognitive behavioral therapy for insomnia, medication adjustment, and adjunctive neuromodulation.

OUTCOMES: During the overall multimodal treatment course, on the night of the first BST session, the patient initiated sleep without alcohol and returned to sleep after a nocturnal awakening without further drinking. Polysomnography showed a total sleep time of 373 minutes and sleep efficiency of 85.6%. Following the second session, he consumed approximately 20 mL of alcohol after a nocturnal awakening before returning to sleep; sleep diary-reported sleep efficiency was 90.6%. Vital signs remained stable during both sessions, and no apparent anesthesia- or sedation-related adverse events occurred. No further alcohol intake was recorded from hospital day 3 to day 7. At discharge, subjective sleep efficiency was 87.7%, with improvements in sleep initiation, maintenance, and some daytime symptoms. Postdischarge follow-up data were unavailable.

LESSONS: BST using propofol and dexmedetomidine may warrant further investigation as an adjunct to multimodal management in patients with chronic insomnia, long-term alcohol use as a sleep aid, and withdrawal-related symptoms. However, improvements cannot be attributed specifically to BST, and its contribution and durability require prospective evaluation.},
}

Humans
Male
Aged
*Sleep Initiation and Maintenance Disorders/therapy
*Substance Withdrawal Syndrome/therapy
Hypnotics and Sedatives/administration & dosage/therapeutic use
Combined Modality Therapy
Dexmedetomidine/administration & dosage/therapeutic use
Propofol/administration & dosage/therapeutic use
Cognitive Behavioral Therapy/methods
*Alcoholism/complications

@article {pmid42855049,
year = {2026},
author = {Shaukat, W and Nobrega, DB and Rehman, A and van Heumen, J and Roos, R and de Pater, M and Liaqat, C and Hall, DC and Lhermie, G and Kastelic, JP and Kelton, DK and De Buck, J and Barkema, HW},
title = {Bayesian estimation of true herd-level prevalence and spatial distribution of Salmonella Dublin-seropositive dairy herds in Alberta, Canada.},
journal = {Journal of dairy science},
volume = {},
number = {},
pages = {},
doi = {10.3168/jds.2026-29221},
pmid = {42855049},
issn = {1525-3198},
abstract = {Salmonella Dublin, a cattle-adapted serovar of Salmonella enterica ssp. enterica with zoonotic potential, has become a pathogen of increasing concern for the dairy industry in Canada and beyond. A comprehensive understanding of its epidemiology is crucial to guide effective surveillance and control. The objectives of this serial cross-sectional study were to estimate the true herd-level seroprevalence of Salmonella Dublin and to characterize the spatial distribution of Salmonella Dublin-seropositive dairy herds in Alberta, Canada. Bulk tank milk (BTM) samples were collected from all active dairy herds in Alberta, Canada, in June (n = 462) and December (n = 451) 2024, and PrioCHECK ELISA was used to detect antibodies against Salmonella Dublin. Herds were classified as positive if ≥ 1 sample had a percent positivity ≥ 35. True seroprevalence was estimated using a Bayesian latent class model accounting for imperfect sensitivity and specificity. Spatial patterns were evaluated using a Bayesian geostatistical model, and a spatial scan statistic was used to detect localized clusters. The estimated true prevalence of Salmonella Dublin-seropositive herds in Alberta was 14.9% (95% Bayesian credible interval: 7.3-28.2%), with regional estimates of 8.0% (95% BCI = 1.6-22.6%), 13.9% (95% BCI = 4.2-31.8%), and 20.9% (95% BCI = 8.5-42.5%) for the north, central, and south regions of Alberta, respectively. The continuous spatial risk surface revealed that the true herd seropositivity varied markedly across Alberta, with some areas having much higher predicted seropositivity probabilities than others. Furthermore, 2 localized clusters of higher seropositivity were detected: a prominent cluster (95-km radius) in southwestern Alberta and a relatively smaller cluster (19-km radius) in a high-density dairy corridor in the central region. These findings indicate that Salmonella Dublin exposure is moderate in Alberta, with marked regional heterogeneity and pronounced spatial gradients in herd-level seropositivity. Incorporating spatial context into surveillance and control may improve targeting of interventions aimed at reducing within-region transmission. These findings also illustrate the value of Bayesian latent class modeling for generating bias-adjusted true seroprevalence estimates in the absence of a gold-standard diagnostic test.},
}

@article {pmid42855606,
year = {2026},
author = {Jahanian Najafabadi, A and Longo, MR and Liepelt, R},
title = {The dynamic bodily self across lifespan: Agency, ownership, and the plasticity of internal body models.},
journal = {Psychonomic bulletin & review},
volume = {33},
number = {8},
pages = {},
pmid = {42855606},
issn = {1531-5320},
mesh = {Humans ; *Sense of Agency ; *Body Image ; *Illusions/physiology ; *Aging/physiology ; },
abstract = {Two fundamental aspects of research on bodily self-consciousness, agency and body ownership, have recently seen exponential growth. The seminal paper of Botvinick and Cohen in 1998 provided a vital psychophysical tool for studying the bodily self - the rubber hand illusion, in which one feels as if an external object becomes a part of one's body. Recent motor control views of ownership highlight the influence of agency, the sense that one is in control of one's movements. Technological advancements have facilitated more ecologically valid experimental paradigms, such as virtual versions of the rubber hand illusion, furthering our understanding of the mechanisms through which the brain generates agency and ownership sensations. Recent findings suggest that agency and ownership can interact while remaining dissociable, as one may experience agency over an object that is not embodied or ownership of an object that cannot be controlled. Understanding the neurocognitive mechanisms underlying agency and ownership is critical for addressing the challenges of healthy aging. As sensory-motor precision declines across the lifespan, maintaining a coherent sense of body and control becomes essential for mobility, the successful adoption of assistive technologies, and overall cognitive resilience. This review synthesizes current findings on the agency-ownership interaction, discusses its neurobiological substrates, and explores its transformative potential for the design of "self-inclusive" prosthetics and brain-machine interfaces. We conclude by outlining a framework showing how optimization of these processes can support functional independence and psychological well-being in an aging population.},
}

Humans
*Sense of Agency
*Body Image
*Illusions/physiology
*Aging/physiology

@article {pmid42855610,
year = {2026},
author = {Shibuya, J and Kyoda, Y and Hashimoto, K and Kobayashi, K and Tanaka, T and Masumori, N},
title = {Changes in Bladder Outlet Obstruction After Photoselective Vaporization of the Prostate in Men With and Without Detrusor Underactivity: A Retrospective Propensity Score-Matched Study.},
journal = {Lower urinary tract symptoms},
volume = {18},
number = {6},
pages = {e70100},
doi = {10.1111/luts.70100},
pmid = {42855610},
issn = {1757-5672},
mesh = {Humans ; *Urinary Bladder Neck Obstruction/surgery/physiopathology/etiology ; Male ; Retrospective Studies ; *Prostatic Hyperplasia/surgery/complications ; *Urinary Bladder, Underactive/complications/physiopathology ; Aged ; Urodynamics ; Propensity Score ; *Laser Therapy/methods ; Treatment Outcome ; Middle Aged ; },
abstract = {OBJECTIVES: To compare symptomatic improvement after photoselective vaporization of the prostate (PVP) between patients with and without detrusor underactivity (DU), together with objective changes in urodynamic parameters assessed by pressure-flow study (PFS).

METHODS: This single-center retrospective study evaluated 136 benign prostatic hyperplasia (BPH) patients treated with a 180 W PVP-XPS system from 2020 through 2025. The patients were divided into DU and non-DU groups based on the bladder contractility index (BCI), with DU defined as BCI < 100. We retrospectively analyzed selected patients with BPH and clinically or urodynamically relevant bladder outlet obstruction (BOO) who underwent PVP and completed PFS before and 6 months after surgery. Propensity score matching (PSM) was applied to adjust for age and the baseline BOO index (BOOI).

RESULTS: A total of 105 patients were included in the final analysis and 53 were in the DU group. The DU group had a significantly lower BOOI (48.6 ± 17.5 vs. 71.4 ± 32.3, p < 0.01) than the non-DU group. After PSM, age and baseline BOOI were well balanced, but BCI remained significantly different. No significant differences between the two groups were observed in questionnaire scores or uroflowmetry findings at 6 months after surgery. The BOOI at 6 months after surgery was not significantly different between the DU group (13.4 ± 13.7) and the non-DU group (13.7 ± 19.3) (p = 0.93). BOOI reduction was also not significantly different between the DU group (-37.0 ± 24.1) and the non-DU group (-36.2 ± 24.5) (p = 0.90).

CONCLUSION: In selected patients with clinically or urodynamically relevant BOO, significant improvements in BOOI and LUTS were observed in patients with DU after PVP, with no statistically significant between-group differences detected at 6 months.},
}

Humans
*Urinary Bladder Neck Obstruction/surgery/physiopathology/etiology
Male
Retrospective Studies
*Prostatic Hyperplasia/surgery/complications
*Urinary Bladder, Underactive/complications/physiopathology
Aged
Urodynamics
Propensity Score
*Laser Therapy/methods
Treatment Outcome
Middle Aged

@article {pmid42855737,
year = {2026},
author = {La, Y and Yuan, J and Chen, J and Lv, K and Wang, X and Gao, Q and Wang, Y and Wang, C and Xu, B and Yang, Y and Zhang, J},
title = {The glymphatic-lymphatic axis: known pathways, open questions, and emerging roles in neurodegeneration.},
journal = {Translational neurodegeneration},
volume = {15},
number = {1},
pages = {},
pmid = {42855737},
issn = {2047-9158},
support = {LY24H090006//Natural Science Foundation of Zhejiang Province/ ; 82571418//National Natural Science Foundation of China/ ; 32530027//National Natural Science Foundation of China/ ; 2024C03098//Key Research and Development Program of Zhejiang Province/ ; },
mesh = {Humans ; *Glymphatic System/metabolism/pathology ; *Neurodegenerative Diseases/pathology/metabolism/physiopathology ; Animals ; *Lymphatic Vessels/metabolism/pathology ; *Brain/metabolism ; Lymphatic System ; Cerebrospinal Fluid/metabolism ; },
abstract = {The discovery of the glymphatic system and meningeal lymphatic vessels has transformed our understanding of brain fluid homeostasis and central-peripheral communication, highlighting the critical role of cerebrospinal fluid in brain clearance. While substantial progress has been made in defining the structural components and physiological regulators of brain clearance, major uncertainties remain regarding (1) how these systems interact as an integrated physiological network, (2) how they fail in disease, and (3) whether they can be therapeutically manipulated. In this review, we first summarize the anatomical and physiological organization of the glymphatic-lymphatic axis, with particular emphasis on its regulatory mechanisms in light of rapid advances in this field. We then discuss how the glymphatic-lymphatic axis functionally integrates with peripheral immune and metabolic systems and describe how impairment of these clearance pathways is associated with the accumulation of pathological proteins, neuroinflammation, and vascular dysfunction and may contribute to the progression of neurodegenerative diseases such as Alzheimer's and Parkinson's diseases. Finally, we discuss emerging therapeutic strategies and biomarkers targeting the glymphatic-lymphatic axis. By redefining the central-peripheral crosstalk, the glymphatic-lymphatic axis provides a unifying framework for understanding brain clearance, neurodegeneration, and future therapeutic interventions.},
}

Humans
*Glymphatic System/metabolism/pathology
*Neurodegenerative Diseases/pathology/metabolism/physiopathology
Animals
*Lymphatic Vessels/metabolism/pathology
*Brain/metabolism
Lymphatic System
Cerebrospinal Fluid/metabolism

@article {pmid42850796,
year = {2026},
author = {Zheng, X and Araújo, J and Busson, Q and Goswami, U},
title = {Towards a Brain-Computer Interface (BCI) for Improving Phonological Processing in Developmental Dyslexia: An Exploratory Study.},
journal = {The European journal of neuroscience},
volume = {64},
number = {7},
pages = {e70709},
pmid = {42850796},
issn = {1460-9568},
support = {//Yidan Prize Foundation/ ; },
mesh = {Humans ; *Dyslexia/physiopathology ; *Brain-Computer Interfaces ; Female ; Electroencephalography/methods ; Male ; Reading ; Adult ; *Speech Perception/physiology ; *Brain/physiopathology ; Young Adult ; Phonetics ; },
abstract = {Brain-computer interfaces (BCIs) have immense potential regarding the provision of therapies for disorders of development, but to date have typically been created for non-linguistic disorders such as ADHD (attention deficit hyperactivity disorder). Here, we present a BCI designed to improve linguistic phonological processing during natural speech listening, aimed at participants with developmental dyslexia. Phonological 'deficits' are considered a core feature of dyslexia across languages and are present from infancy in those later diagnosed with dyslexia. Here, a non-invasive EEG-BCI relying on auditory inputs and visual feedback was developed to optimise brain patterns related to phonological development in infants and children. The BCI focused on neural patterns identified using Temporal Sampling theory, which has found the theta/delta ratio during natural speech listening to be a key neural associate of impaired phonology. As a first step, the BCI was tested with adults. Adults with and without dyslexia played the BCI for 16 sessions and received pre- and post-testing regarding phonological awareness and single word and non-word reading skills. Learning was assessed both online, as in most BCI studies, and offline, enabling removal of a greater number of potential EEG artefacts. Significant associations between offline BCI scores (a measure of BCI learning) and improvements in non-word reading were found for all participants who learned the BCI, both dyslexic and control. BCI learning also showed significant associations with improvements in real word reading and amplitude rise time discrimination. The data are interpreted with respect to Temporal Sampling theory.},
}

Humans
*Dyslexia/physiopathology
*Brain-Computer Interfaces
Female
Electroencephalography/methods
Male
Reading
Adult
*Speech Perception/physiology
*Brain/physiopathology
Young Adult
Phonetics

@article {pmid42850844,
year = {2026},
author = {Annikov, YG and Chekhonatskiy, AA and Komleva, NE and Chekhonatskiy, VA and Urzaleeva, IA and Annikova, OV and Mazilov, SI},
title = {[THE NEUROPROSTHETICS AND NEUROSTIMULATION: HISTORY OF DEVELOPMENT, KEY ACHIEVEMENTS AND PROSPECTS (THE REVIEW)].},
journal = {Problemy sotsial'noi gigieny, zdravookhraneniia i istorii meditsiny},
volume = {34},
number = {4},
pages = {825-833},
doi = {10.32687/0869-866X-2026-34-4-825-833},
pmid = {42850844},
issn = {0869-866X},
mesh = {Humans ; *Brain-Computer Interfaces/history ; History, 20th Century ; History, 21st Century ; },
abstract = {The purpose of the study is to systematize data from National and foreign publications on history of development, key achievements and prospects of application of neuroprosthetics and neurostimulation. The analysis was carried out concerning databases of PubMed, eLibrary, Cohrane Library, MEDLINE over 2000-2025 according keywords neuroprosthetics, neuroengineering, cognitive neuroprosthesis, brain-computer interface, neurostimulation. The review includes 68 full-text sources.},
}

Humans
*Brain-Computer Interfaces/history
History, 20th Century
History, 21st Century

@article {pmid42851647,
year = {2026},
author = {Li, J and Sha, L and Yan, X and Wang, C and Shi, J and Qiao, H and Wang, C and Wei, Y and Gong, C},
title = {PADA: perturbation-aware differential Riemannian attention for robust motor imagery decoding.},
journal = {Frontiers in neuroscience},
volume = {20},
number = {},
pages = {1943934},
pmid = {42851647},
issn = {1662-4548},
abstract = {INTRODUCTION: Riemannian geometry models have significantly improved the robustness and performance of EEG-based motor imagery decoding and are better suited to small sample sizes; however, their ability to extract nonlinear and spatiotemporal variability features often falls short of Euclidean-based deep learning methods. To address these limitations, we introduce a perturbation-aware differential Riemannian attention mechanism (PADA).

METHODS: PADA develops a window-based learnable Riemannian module that maps informative segments onto the manifold to characterize nonlinear features while preserving local structure, together with a two-stream differential manifold attention mechanism. Tangent-space perturbations are used as a probe for feature instability, and attention weights that are hypersensitive to noise are differentially suppressed to distill robust geometric dependencies.

RESULTS: On the BCI Competition IV-2a and Zhou2016 datasets, PADA achieved competitive clean-condition accuracies of 81.44% and 92.79%, respectively. Under 50% noise, its accuracy dropped by only 12%, achieving the best classification performance among the compared models. On a self-constructed dataset involving 227 subjects, PADA achieved up to 70.46% binary classification accuracy on the classic hand motor imagery task using only 12 training samples.

DISCUSSION: These results indicate that PADA provides robust and effective feature learning for noisy and data-limited brain-computer interface applications.},
}

@article {pmid42848345,
year = {2026},
author = {Lu, M and Di, J and Deng, M and Tang, Z and He, J and Lu, Z and Liu, B},
title = {Single-cell and Spatial Omics Reveal Spinal Cord Injury Microenvironmental Heterogeneity Relevant to Neural Interfaces and Spinal Neuromodulation.},
journal = {Neuromodulation : journal of the International Neuromodulation Society},
volume = {},
number = {},
pages = {},
doi = {10.1016/j.neurom.2026.08.010},
pmid = {42848345},
issn = {1525-1403},
abstract = {OBJECTIVE: Spinal cord injury (SCI) interrupts brain-spinal communication and causes persistent motor, sensory, and autonomic deficits.

MATERIALS AND METHODS: We summarize the current clinical and translational evidence on brain-computer interface (BCI)/epidural electrical stimulation (EES)/brain-spinal interface (BSI) for locomotor recovery after traumatic SCI and outline the major patterns of microenvironmental heterogeneity revealed by single-cell and spatial omics. We then link these patterns to clinically relevant axes-acute vs subacute versus chronic phases, complete vs incomplete injury, and peri-lesional vs scar-dominated regions-and discuss how they may influence circuit excitability, plasticity, and responsiveness to neuromodulation.

RESULTS: BCIs/EESs/BSIs have enabled standing, stepping, or walking-related functions, but clinical outcomes remain highly variable. Emerging single-cell and spatial omics studies indicate that the injured spinal cord is not a uniform lesion but consists of distinct microenvironments that differ across species, anatomical regions, cell states, and stages after injury. We further propose a staged, microenvironment-aware framework linking injury stage, tissue state, and neuromodulation response through testable hypotheses. Representative clinical scenarios are presented as conceptual models rather than as prescriptive algorithms.

CONCLUSIONS: Although these relationships remain hypothesis-generating, integrating neuromodulation with omics-based profiling may provide a foundation for developing more biologically informed and individually adapted strategies for SCI rehabilitation.},
}

@article {pmid42849154,
year = {2026},
author = {Zhang, Y and Piao, Q and Yang, Y and Boland, JE and Du, Y and Wang, Y and Han, M},
title = {Corrigendum to "When Chinese verb transitivity meets wrong syntactic category" [Brain Lang. 281 (2026) 105820].},
journal = {Brain and language},
volume = {283},
number = {},
pages = {105849},
doi = {10.1016/j.bandl.2026.105849},
pmid = {42849154},
issn = {1090-2155},
}

@article {pmid42849701,
year = {2026},
author = {Sharma, N and Kumar, S and Prakash, C},
title = {A comparative study and analysis of recent EEG-based brain-computer interface approaches and their applications.},
journal = {Neuroscience},
volume = {},
number = {},
pages = {},
doi = {10.1016/j.neuroscience.2026.10.006},
pmid = {42849701},
issn = {1873-7544},
abstract = {Brain-Computer Interface (BCI) is an innovative tool that interprets brain instructions and enables humans for using their mental instructions to operate computers, robots, and other external equipment. People with motor disabilities may now utilize external devices much more easily because to recent advancements in BCI technology. This study analyzes newly developed models and provides a comprehensive literature overview of current developments for the field of completely EEG-based BCI systems. In this paper, the basic concepts of EEG-based BCIs are presented, and the key constituents of the system, such as electrophysiological control signals, feature extraction, classification, and performance measurements in different BCI applications, are discussed. The paper considers the brain's functional connectivity, which is studied through relations among different brain areas as connectivity patterns that reflect information exchange between them at rest and during certain tasks. A comparison among various deep learning models for EEG-Based BCI with subject dependent mode is also presented. The result shows that ATCNet model gives highest accuracy with 86.46% and the lowest accuracy was given by EEG-TCNet with 50.69.},
}

@article {pmid42850277,
year = {2026},
author = {Wang, C and Liu, Z and Zhang, W and Yang, R and Li, Y and Wang, Z and Chu, D and Zhou, X and Hong, W and Wang, S and Wang, X and Wang, J and Li, X and Zhang, W and Wang, G and Lu, S and Yin, J and Wu, M and Liu, J},
title = {A freestanding submicron electrocorticography electrode array.},
journal = {Nature communications},
volume = {17},
number = {1},
pages = {},
pmid = {42850277},
issn = {2041-1723},
mesh = {Animals ; *Electrocorticography/instrumentation/methods ; Male ; Electrodes, Implanted ; Rats ; Epilepsy/physiopathology ; *Brain-Computer Interfaces ; Cerebral Cortex/physiology ; Signal-To-Noise Ratio ; Electrodes ; Odonata ; },
abstract = {Electrocorticography electrode arrays (EEAs) are key brain-computer interfaces. To be freestanding, which lays the foundation of fabrication and implantation, EEAs need to maintain sufficient rigidity. However, to form seamless and damage-free contact on the cerebral cortex, EEAs should be super-flexible, raising a seemingly unsolvable dilemma. Here, inspired by dragonfly wings, we present a freestanding submicron EEA (fs-EEA) that combines rigidity and flexibility, which are typically mutually exclusive characteristics. A wing vein-like frame is introduced to enable the freestanding capability, while the wing membrane-like functional region of fs-EEA is reduced to 850-nm-thickness that boosts its softness. This fs-EEA ensures excellent cortical conformality and biocompatibility on male epileptic rat models, improves electrocorticographic amplitude and signal-to-noise ratio. Our vein-membrane hybrid structure strategy thus resolves the inherent contradiction of EEAs, and has the potential to support advances in brain science and related fields.},
}

Animals
*Electrocorticography/instrumentation/methods
Male
Electrodes, Implanted
Rats
Epilepsy/physiopathology
*Brain-Computer Interfaces
Cerebral Cortex/physiology
Signal-To-Noise Ratio
Electrodes
Odonata

@article {pmid42843403,
year = {2026},
author = {Lee, S},
title = {Ordered onset of regional desynchronization and a transition in movement-class discriminability during cue-paced motor imagery.},
journal = {Journal of neural engineering},
volume = {},
number = {},
pages = {},
doi = {10.1088/1741-2552/aeb161},
pmid = {42843403},
issn = {1741-2552},
abstract = {Motor imagery recruits a distributed sensorimotor network, yet the order in which its regions are engaged has not been tested statistically within the healthy brain, and the interval in which information about the imagined movement appears remains unidentified. Early post-cue classification peaks have been attributed to visually induced orienting rather than to imagery, an inference drawn from a condition contrast rather than direct measurement.Two independent public datasets were analysed under a matched pipeline of causal filtering, ocular regression and current source density: Weibo2014 as the primary dataset and BCI Competition IV dataset 2a for validation. Mu and beta desynchronization onset was ordered across cortex, from occipital through sensorimotor and midline pre-supplementary motor area to lateral prefrontal. The ordering held under all nine onset criteria examined, under channel-count matching and within-trial pairing, and after removal of the sensorimotor component. Class-discriminative information rose most steeply between 0.60 and 0.75 s and saturated within a further 0.3 s. This transition lay closer to sensorimotor onset than to cue presentation in every participant and fell at or after that onset in nearly all of them, although the two were not shown equivalent at the resolution of the decoding grid. Windows centred on each region's own onset classified equivalently to reassigned windows, so the ordering serves as a temporal landmark rather than a decodable feature, locating the interval in which information about the imagined movement becomes available.},
}

@article {pmid42845593,
year = {2026},
author = {Yi, J and Patel, KK and Miller, RJH and Marcinkiewicz, AM and Kamagate, A and Shanbhag, A and Hijazi, W and Lemley, M and Zhou, J and Liang, JX and Ramirez, G and Mostafavi, S and Urs, M and Spielvogel, CP and Slipczuk, L and Travin, M and Alexanderson, E and Caraval-Juarez, I and Packard, RRS and Al-Mallah, M and Ruddy, TD and Einstein, AJ and Feher, A and Miller, EJ and Acampa, W and Knight, S and Le, VT and Mason, S and Calsavara, VF and Chareonthaitawee, P and Wopperer, S and Kwan, AC and Wang, L and Li, D and Fishman, EK and Lopez-Ramirez, F and Berman, DS and Kwiecinski, J and Dey, D and Carli, MD and Slomka, PJ},
title = {Beyond BMI: an interpretable integrated body composition index from low-dose chest CT for all-cause mortality risk stratification: a multicentre study.},
journal = {medRxiv : the preprint server for health sciences},
volume = {},
number = {},
pages = {},
doi = {10.64898/2026.08.05.26359437},
pmid = {42845593},
abstract = {BACKGROUND: Body composition is recognized as a major determinant of health outcomes, but its multidimensional nature makes clinical adoption challenging. We sought to develop and validate a body composition index (BCI) for all-cause mortality risk assessment, integrating variables of six body composition tissues.

METHODS: We analyzed 28509 consecutive patients undergoing myocardial perfusion imaging with routine low-dose chest CT attenuation correction (CTAC) scans acquired during myocardial perfusion imaging (MPI) at 12 centers across four countries. An artificial intelligence-based BCI was developed in a cohort of 15037 patients' CTACs by integrating the CT-derived metrics of bone, skeletal muscle, and four adipose tissue compartments, coronary artery calcium score, and basic demographic variables (age, sex, BMI). The performance of BCI for mortality prediction was validated in an internal cohort of 6444 patients and an external cohort of 7028 patients by prognosis, calibration, net benefit, and explainability. Model-based simulation of tissue metrics modification was performed to evaluate estimated mortality risk reduction.

FINDINGS: During a median of 3.5 (IQR [1.9, 5.1]) years, 4697 (16%) patients died. In the external testing cohort, the BCI demonstrated excellent discrimination for mortality (area under receiver operating characteristic curve 0.78 (95% CI [0.76, 0.79]) and Harrell's concordance index 0.75 [0.73, 0.76]), calibration, and net benefit overall and across pre-specified subgroups stratified by patient characteristics and imaging protocols. Visceral adipose tissue attenuation was the most influential body composition measure, followed by skeletal muscle volume. Simulated improvement in body composition was associated with significant mortality risk reduction.

INTERPRETATION: An index combining six body composition measures obtained opportunistically from routine chest CT provides robust mortality risk stratification. By converting complex body composition information into a single interpretable score, the BCI can facilitate clinical implementation of opportunistic CT biomarkers and guide individualized preventive strategies.

FUNDING: National Institute of Health.

RESEARCH IN CONTEXT: Evidence before this study: We searched PubMed and Google Scholar on July 7, 2026 for English language studies using the terms ("mortality prediction" OR "death prediction") AND ("integrated body composition analysis" OR "body composition metrics integration") AND ("predictive model" OR "model development"), and identified 41 publications describing predictive models incorporating body composition metrics.Recent advances in artificial intelligence (AI) have enabled increasingly granular quantification of body composition from CT, with multiple tissues and tissue-specific metrics showing independent prognostic value for all-cause mortality (ACM) prediction. However, reported associations of body composition metrics with mortality risk were highly inconsistent or even contradictory across studies. This heterogeneity likely reflected the differences in study design or the inherent complexity of body composition analysis, such as the interaction among variables derived from multiple tissues. Thus, it is difficult to interpret body composition results reliably in clinical practice. We hypothesized that integrating multiple body composition metrics could achieve comprehensive and reliable prediction of mortality across patient populations.Added value of this study: In this large international, multicenter longitudinal cohort study of 28509 consecutive patients with suspected or known coronary artery disease (CAD) undergoing myocardial perfusion imaging across 12 sites in four countries, we developed and externally validated an artificial intelligence-based body composition index (BCI) for all-cause mortality prediction. The BCI integrated chest CT-derived metrics of six body composition tissue compartments - bone, skeletal muscle (SM), subcutaneous adipose tissue (SAT), intramuscular adipose tissue (IMAT), visceral adipose tissue (VAT), and epicardial adipose tissue (EAT), together with demographics (age, sex, BMI) and CT-derived coronary artery calcium (CAC) score, all derived automatically from routine low-dose attenuation correction CT scans without additional imaging or laboratory tests.In an external validation cohort of 7028 patients, the BCI achieved robust discrimination for mortality prediction, substantially outperforming a baseline model using demographics and calcium score alone, and exceeded the discriminative performance of prior body composition models. Performance was consistent across subgroups defined by sex, age, body mass index, cardiometabolic risk factors, imaging modality, and acquisition protocol, supporting generalizability across clinical settings. The BCI demonstrated excellent calibration and net clinical benefit. Explainability analysis identified visceral adipose tissue attenuation and skeletal muscle volume as the dominant contributors to mortality risk, offering tissue-specific therapeutic targets that complement generic weight-based metrics. Model-based simulations suggested that improvements in these specific compartments could reduce estimated mortality risk more effectively than equivalent changes in BMI alone, pointing toward precision therapeutic targets beyond weight reduction.Our study improved over previous literatures on body composition-based predictive models in one or more of the following ways: (1) we considered six types of body composition tissues while existing studies considered only three or less tissues, thus significantly higher granularity and improved ability to capture refined tissue interactions; (2) our model was rigorously validated in external sites, showing cross-site generalizability while most previous work lacked external validation; (3) our model achieved significant prediction performance improvement over previous body composition predictive models, offering better clinical outcomes prediction and facilitating the suitability for widespread clinical adoption; (4) our study was conducted with a large multi-center cohort across multiple countries, achieving statistical power and demonstrating reliability towards patients characteristics and imaging acquisition protocols while previous studies involved small single-center cohorts; (5) our model required only CT scans and basic demographics which were widely available in clinical routine while the existing models required dedicated laboratory tests or specialized imaging modalities, enhancing the applicability in routine clinical practice.Implications of all the available evidence: BCI offers an explainable and clinically deployable tool for improved prediction of mortality risk from existing chest CT scans. Its use could inform updates to existing risk models, guide precision preventive strategies, and support development of therapeutics targeting specific body composition phenotypes.},
}

@article {pmid42845720,
year = {2026},
author = {Olivares, CY and Schettini, N},
title = {A high-precision synchronization and centroid-based framework for single-trial decoding of perceived natural speech.},
journal = {Frontiers in computational neuroscience},
volume = {20},
number = {},
pages = {1947069},
pmid = {42845720},
issn = {1662-5188},
abstract = {Decoding natural speech from EEG is constrained by low signal-to-noise ratios, commonly addressed through event-related potential (ERP) grand averaging. However, this approach acts as a low-pass filter, obscuring trial-to-trial variability and high-frequency neural dynamics. To overcome these limitations, we propose an open-box, FPGA-synchronized framework for single-trial EEG analysis of Spanish vowels (/a/, /i/, /u/). The proposed architecture enforces sub-millisecond hardware determinism, eliminating software-induced temporal jitter and enabling robust extraction of high-frequency oscillatory features, including the Gamma band, without relying on averaged ERPs. This synchronization framework facilitates the separation of early exogenous sensory responses from late endogenous cognitive processing, while comprehensive ablation studies were performed to evaluate the effect of software-induced jitter on decoding performance. Results showed that introducing simulated jitter degraded classification performance to theoretical chance levels (33.20%), confirming hardware synchronization as a strict prerequisite for reliable single-trial decoding. The framework was further validated using a Centroid-PCA-SVM pipeline operating on a 16-channel spatial subset. Using the full proposed architecture, a grand mean classification accuracy of 67.70% was achieved, substantially exceeding the theoretical chance level of 33.33%. The across-participant mean of the maximum fold accuracies reached 83.00%, with individual validation folds achieving up to 90.00%. In contrast, zero-calibration leave-one-subject-out (LOSO) generalization remained at chance level (33.08%; κpooled = -0.0036), indicating limited cross-participant transferability and highlighting the current need for participant-specific calibration or alignment. Overall, these findings demonstrate that hardware-deterministic EEG acquisition enables the recovery of robust trial-by-trial brain dynamics that are typically obscured by conventional averaging approaches. By combining precise synchronization with low-latency algorithmic inference, the proposed framework establishes a foundation for real-time neural decoding and represents a stepping stone for practical brain-computer interfaces and advanced clinical monitoring applications.},
}

@article {pmid42845760,
year = {2026},
author = {Hai, X and Shen, R and Wang, X and Yang, GM},
title = {Noninvasive Intelligent Brain-Metasurface System for Real-Time Electromagnetic Wave Manipulation.},
journal = {Nanophotonics (Berlin, Germany)},
volume = {15},
number = {13},
pages = {e70194},
pmid = {42845760},
issn = {2192-8614},
abstract = {A noninvasive intelligent brain-computer interface (BCI) is integrated with a reconfigurable metasurface to establish a direct neural-to-electromagnetic transduction pathway for real-time, multimodal wave control. By combining convolutional-neural-network-based decoding of motor-imagery electroencephalography (EEG) signals with threshold-based recognition of electrooculography (EOG) commands, the proposed system converts neural and ocular activities into high-speed control commands, which are subsequently implemented through an FPGA-driven dual-polarized reconfigurable metasurface for real-time electromagnetic-wave manipulation. This architecture concurrently enables beam steering, polarization conversion, and radar cross-section (RCS) reduction within a unified platform. Three distinct operational modes are demonstrated: polarization-invariant beam deflection, joint steering and polarization conversion, and broadband RCS reduction via random phase coding. Experimental results reveal a peak gain of 18.54-19.29 dBi with a 3 dB beamwidth of 12°-15° at a +20° steering angle, and an RCS reduction exceeding 10 dB across 3.7-3.85 GHz. The proposed intelligent BCI-metasurface paradigm validates the feasibility of brain-controlled electromagnetic manipulation, opening new avenues for adaptive wireless communications, immersive virtual environments, and next-generation assistive technologies.},
}

@article {pmid42846039,
year = {2026},
author = {Alanoğlu, M and Khudhur, RS and Demirtaş, Z and Uslukaya, A},
title = {Distributed leadership and teachers' work engagement: subjective well-being as a mediator and perceived collective teacher efficacy as a moderator.},
journal = {Frontiers in psychology},
volume = {17},
number = {},
pages = {1887759},
pmid = {42846039},
issn = {1664-1078},
abstract = {INTRODUCTION: This study examined the association between distributed leadership and teachers' work engagement, with teacher subjective well-being specified as a mediator and perceived collective teacher efficacy as a moderator.

METHODS: Drawing on the Job Demands-Resources model and social cognitive theory, a moderated mediation model was tested using Bayesian structural equation modeling. Data were collected from 598 teachers working in 60 public schools in Erbil, Kurdistan Region of Iraq.

RESULTS: Distributed leadership was positively associated with teacher subjective well-being [β = 0.233, 95% BCI (0.077, 0.381)] and work engagement [β = 0.234, 95% BCI (0.135, 0.333)], while subjective well-being was positively associated with work engagement [β = 0.172, 95% BCI (0.112, 0.232)]. The indirect association through subjective well-being was supported [B = 0.039, 95% BCI (0.012, 0.074)]. Perceived collective teacher efficacy moderated the association between distributed leadership and subjective well-being [β = 0.099, 95% BCI (0.003, 0.192)]. The conditional indirect association was supported at the mean and higher levels of perceived collective teacher efficacy but not at lower levels, and the index of moderated mediation excluded zero [B = 0.019, 95% BCI (0.001, 0.043)]. These conclusions remained stable under alternative prior specifications.

DISCUSSION: The findings indicate an indirect statistical association between distributed leadership and work engagement through teacher subjective well-being, which varied according to perceived collective teacher efficacy. Given the cross-sectional, self-report design, this mediation represents a statistical decomposition of covariance rather than evidence of a causal mechanism or temporal sequence.},
}

@article {pmid42846040,
year = {2026},
author = {Alba Diaz, PA and Kwon, J and Berstis, K and Weed, M and Ferrari, P and Wang, J},
title = {Optimal sensor set for MEG-based spontaneous and intended speech decoding toward practical communication.},
journal = {Frontiers in human neuroscience},
volume = {20},
number = {},
pages = {1925703},
pmid = {42846040},
issn = {1662-5161},
abstract = {INTRODUCTION: Neural speech decoding has demonstrated great potential as an alternative to current brain-computer interface (BCI) spellers, enabling a faster, more natural way of communication for patients who have lost voluntary abilities, including speech. Most neural speech decoding studies have focused on decoding cued speech, in which participants are presented with visual or auditory signals before producing imagined or overt speech. Only a few recent studies have investigated the decoding of spontaneous and intended speech without external cues using neural activity acquired from a full-scalp clinical magnetoencephalography (MEG) system. With recent advances in MEG technology, including optically pumped magnetometers (OPMs) and nitrogen-vacancy (NV) center (NV-diamond) magnetometers, MEG is emerging as a next-generation wearable system for practical speech BCIs. However, it remains unclear how many sensors are needed, where to place them, and what sampling frequency is optimal for wearable MEG devices that require cost-effective, computationally efficient operation.

METHODS: In this study, we addressed these gaps by using a forward selection algorithm to find an optimal subset of sensors for spontaneous intended and overt speech decoding and by evaluating decoding performance across different frequency bands. Support vector machine and random forest were used as the classifiers in these experiments.

RESULTS: Our results indicate that only a small number of sensors (7-8 on average, with a maximum of fewer than 24) are sufficient to achieve decoding performance comparable to that obtained using the whole scalp (more than 200) sensors. In addition, we observed that the Lower High-Gamma (62-125 Hz) band yielded the best performance in some settings, although not statistically significantly higher than other bands in all settings.

DISCUSSION: These findings suggest that future wearable MEG devices may not need the whole-scalp sensor set and should support a sampling rate of at least 250 Hz, while 500 Hz is also encouraged, if possible. In summary, these findings provide practical insights for MEG-based BCIs for communication.},
}

@article {pmid42846320,
year = {2026},
author = {Landeira-Suquia, L and Lee, A and Cuttaz, EA and Goding, J and John, SE and Green, RA},
title = {Endovascular approaches to neural interfacing: lessons from cardiovascular technologies.},
journal = {Frontiers in neuroscience},
volume = {20},
number = {},
pages = {1933040},
pmid = {42846320},
issn = {1662-4548},
abstract = {Endovascular neural interfaces (ENIs) have emerged as a minimally invasive alternative to traditional neural interfaces. These devices are implanted within blood vessels adjacent to neural tissue to record or stimulate neural activity across the vessel wall. This enables minimally invasive implantation through angiographic procedures and reduces the risk of neural tissue damage. As they are deployed in the intravascular space, ENIs have a similar technological evolution to cardiovascular devices and share many features and constraints. First-generation ENIs borrowed guide wires and catheters employed in cardiovascular procedures for endovascular neural recording but were limited to acute settings. Concurrently, the cardiovascular field explored the possibility of integrating electronics onto endovascular devices. Catheters have been used for endovascular stimulation as a treatment for cardiovascular conditions. Cardiovascular stents have been integrated with wireless sensors, enabling self-reporting stents to monitor implant health. In 2016, a breakthrough in the field of ENIs was achieved by leveraging the chronic stability of stents to support an electrode array. Exemplified by Synchron's Stentrode™, stent electrode arrays have been successfully employed as a brain-computer interface for chronic motor cortex recording. Subsequent research in the field has focused on expanding applications and improving spatial resolution. Incorporating valuable lessons from the field of interventional cardiology will be key to accelerate ENI development, especially in the areas of device miniaturisation, enhancement of stimulation capabilities, and wireless communication. Addressing these limitations could revolutionise the treatment of neurological disorders, enabling high spatial resolution recording and stimulation possible through minimally invasive, accessible procedures.},
}

@article {pmid42847828,
year = {2026},
author = {Li, J and Wen, S and Hu, J and Wang, Q and Zhang, B and Zhou, Q and Zhao, J},
title = {Exploiting Phase-Sliding Oscillations for Robust Asynchronous Brain-Computer Interface Control.},
journal = {International journal of neural systems},
volume = {},
number = {},
pages = {2750022},
doi = {10.1142/S0129065727500225},
pmid = {42847828},
issn = {1793-6462},
abstract = {Asynchronous steady-state visual evoked potential (SSVEP)-based brain-computer interfaces (BCIs) show great potential for real-world control, yet their performance is often impaired by phase shifts between ongoing electroencephalogram (EEG) signals and fixed-phase templates. In this study, we observed and validated a phase-sliding oscillation (PSO) phenomenon: when a fixed-phase template is used, phase-sliding trajectories derived from sliding-window sequences with different initial phases exhibit stable periodic oscillations locked to the stimulus frequency. Based on this finding, we propose a PSO-based asynchronous classification (PSO-AC) method. Specifically, EEG signals are reconstructed into sequences with varying initial phases and transformed into stable time-frequency representations via the wavelet synchrosqueezing transform to extract phase-decoupled features. A probabilistic temporal model is then used to characterize temporal dynamics and distinguish control from noncontrol states. Offline experiments (n = 22) showed that PSO-AC achieved a mean control/noncontrol accuracy of 94.2% with a 2[Formula: see text]s data length, outperforming a state-of-the-art baseline. It also maintained superior performance in nine-class classification under different time delays. In online robotic-arm experiments, PSO-AC enabled more stable command triggering and higher control efficiency, reducing the command cost per successful trial from 4.30 to 1.14.},
}

@article {pmid42834306,
year = {2026},
author = {Mhande, Z and Mihale, M and Hellar-Kihampa, H and Brion, N and Baeyens, W},
title = {Isotopic fingerprinting of sediment organic matter reveals pervasive sewage influence in urbanised tropical rivers.},
journal = {Environmental science and pollution research international},
volume = {},
number = {},
pages = {},
pmid = {42834306},
issn = {1614-7499},
support = {296 (ZEIN2014Z163)//VLIR-UOS/ ; },
abstract = {Rapid urbanisation in tropical coastal regions is altering riverine biogeochemistry, yet specific drivers of sediment organic matter (OM) dynamics remain understudied in Sub-Saharan Africa. This study assessed spatial and seasonal variation in sediment total nitrogen (TN), total organic carbon (TOC), molar C/N ratio, and stable isotopes (δ[13]C and δ[15]N) in three coastal rivers draining Dar es Salaam, Tanzania (Kizinga, Mbezi, and Msimbazi). Across all rivers, TN (2.18-5.18 mg/g) and TOC (24.47-50.31 mg/g) concentrations showed no statistically significant seasonal differences (p > 0.05), indicating persistent anthropogenic loading across contrasting hydrological seasons. Msimbazi River exhibited the highest organic enrichment and lowest molar C/N ratios (9.5 ± 1.5-10.5 ± 1.0), consistent with N-rich organic inputs. In contrast, the Kizinga River displayed higher molar C/N ratios but the most enriched δ[15]N values (up to 10.3 ± 1.0‰), consistent with wastewater-derived nitrogen subjected to fractionating transformations such as ammonia volatilisation and denitrification. Bayesian source apportionment using the Food Reconstruction using Isotope Transferred Signals (FRUITS) model estimated that sewage-derived OM contributes a posterior median of 41% (95% Bayesian credible interval (BCI) 28-52%) in Msimbazi and 43% (95% BCI 30-54%) in Kizinga sediments. These findings demonstrate a state of biogeochemical stationarity in which chronic urban wastewater inputs dominate sediment composition across contrasting hydrological seasons. The results highlight that seasonal flushing is insufficient to remove accumulated anthropogenic signatures from riverbeds and underscore the need for structural interventions to reduce sewage inputs to urban tropical rivers.},
}

@article {pmid42835206,
year = {2026},
author = {Li, M and Pan, J and Han, J and Bao, X and Wu, H and Wang, F and Yu, Y and Li, A and Liu, Y and Chen, D and Yang, X and He, Y and Qin, P and Li, Y},
title = {Detecting EEG index of arousal in sleeping patients with unresponsive wakefulness syndrome.},
journal = {iScience},
volume = {29},
number = {10},
pages = {117448},
pmid = {42835206},
issn = {2589-0042},
abstract = {Consciousness is assumed to be defined by two components: awareness and arousal. However, fewer electroencephalographic (EEG) indices have been identified for arousal than for awareness. Since patients clinically diagnosed with unresponsive wakefulness syndrome (UWS) show preserved sleep-wake cycles but markedly reduced reproducible behavioral evidence of awareness, their sleep-wake patterns may provide a useful model for examining arousal-related EEG fluctuations. This study recorded nighttime polysomnography in patients with UWS, patients in a minimally conscious state, and healthy controls. The EEG signal showed that spectral slope could index arousal, fluctuating among sleep stages in all three groups. Both spectral entropy and Lempel-Ziv complexity indexed awareness, showing a significant difference between conscious and unconscious states. These findings provide fundamental evidence for the two-component hypothesis of consciousness.},
}