Brain computer interfaces, a review

Partial face visibility and facial cognition: event-related potential and eye tracking investigation

Face masks became a part of everyday life during the SARS-CoV-2 pandemic. Previous studies showed that the face cognition mechanism involves holistic face processing, and the absence of face features could lower the cognition ability. This is opposed to the experience during the pandemic, when people could correctly recognize faces, although the mask covered a part of the face. This paper clarifies the partial face cognition mechanism of the full and partial faces based on the electroencephalogram (EEG) and eye-tracking data. We observed two event-related potentials, P3a in the frontal lobe and P3b in the parietal lobe, as subcomponents of P300. The amplitude of both P3a and P3b were lowered when the eyes were invisible, and the amplitude of P3a evoked by the nose covered was larger than the full face. The eye-tracking data showed that 16 out of 18 participants focused on the eyes associated with the EEG results. Our results demonstrate that the eyes are the most crucial feature of facial cognition. Moreover, the face with the nose covered might enhance cognition ability due to the visual working memory capacity. Our experiment also shows the possibility of people recognizing faces using both holistic and structural face processing. In addition, we calculated canonical correlation using the P300 and the total fixation duration of the eye-tracking data. The results show high correlation in the cognition of the full face and the face and nose covered (R c = 0.93 ) which resembles the masked face. The finding suggests that people can recognize the masked face as well as the full face in similar cognition patterns.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11571-025-10231-3.

Data alignment based adversarial defense benchmark for EEG-based BCIs

Machine learning has been extensively applied to signal decoding in electroencephalogram (EEG)-based brain-computer interfaces (BCIs). While most studies have focused on enhancing the accuracy of EEG-based BCIs, more attention should be given to their security. Recent findings reveal that EEG-based BCIs are vulnerable to adversarial attacks. To address this, we present the first adversarial defense benchmark based on data alignment, aiming to enhance both the accuracy and robustness of EEG-based BCIs. This study evaluates nine adversarial defense approaches (including five defense strategies) across five EEG datasets (covering three paradigms), three neural networks, and four experimental scenarios. Our results show for the first time that integrating data augmentation, data alignment, and robust training can further improve both the accuracy and robustness of BCIs compared to using only one or two of them. Furthermore, we provide insights into the characteristics of various adversarial defense approaches based on EEG data alignment, offering valuable guidance for developing more accurate and secure EEG-based BCIs.

Filter bank common spatial pattern and envelope-based features in multimodal EEG-fTCD brain-computer interfaces

Brain-computer interfaces (BCIs) provide alternative means of communication and control for individuals with severe motor or speech impairments. Multimodal BCIs have been introduced recently to enhance the performance of BCIs utilizing single modality. In this paper, we aim to advance the state of the art in multimodal BCIs combining Electroencephalography (EEG) and functional transcranial Doppler ultrasound (fTCD) by introducing advanced analysis approaches that enhance system performance. Our EEG-fTCD BCIs employ two distinct paradigms to infer user intent: motor imagery (MI) and flickering mental rotation (MR)/word generation (WG) paradigms. In the MI paradigm, we introduce the use of Filter Bank Common Spatial Pattern (FBCSP) for the first time in an EEG-fTCD BCI, while in the flickering MR/WG paradigm, we extend FBCSP application to non-motor imagery tasks. Additionally, we extract previously unexplored time-series features from the envelope of fTCD signals, leveraging richer information from cerebral blood flow dynamics. Furthermore, we employ a Bayesian fusion framework that allows EEG and fTCD to contribute unequally to decision-making. The multimodal EEG-fTCD system achieved high classification accuracies across tasks in both paradigms. In the MI paradigm, accuracies of 94.53%, 94.9%, and 96.29% were achieved for left arm MI vs. baseline, right arm MI vs. baseline, and right arm MI vs. left arm MI, respectively - outperforming EEG-only accuracy by 3.87%, 3.80%, and 5.81%, respectively. In the MR/WG paradigm, the system achieved 95.27%, 85.93%, and 96.97% for MR vs. baseline, WG vs. baseline, and MR vs. WG, respectively, showing accuracy improvements of 2.28%, 4.95%, and 1.56%, respectively compared to EEG-only results. Overall, the proposed analysis approach improved classification accuracy for 5 out of 6 binary classification problems within the MI and MR/WG paradigms, with gains ranging from 0.64% to 9% compared to our previous EEG-fTCD studies. Additionally, our results demonstrate that EEG-fTCD BCIs with the proposed analysis techniques outperform multimodal EEG-fNIRS BCIs in both accuracy and speed, improving classification performance by 2.7% to 24.7% and reducing trial durations by 2-38 seconds. These findings highlight the potential of the proposed approach to advance assistive technologies and improve patient quality of life.

Unravelling the Brain Resilience Following Stroke: From injury to rewiring of the brain through pathway activation, drug targets, and therapeutic interventions

Synaptic plasticity is a neuron's intrinsic ability to make new connections throughout life. The morphology and function of synapses are highly susceptible to any pathological condition. Ischemic stroke is a cerebrovascular event that affects various brain regions, resulting in the loss of neural networks. Stroke can alter both structural and functional plasticity of synapses, leading to long-term functional disability. Upon ischemic insult, numerous glutamate-mediated synaptic destruction pathways and glial-mediated phagocytic activity are triggered, resulting in excessive synapse loss, altering synaptic plasticity. The conventional stroke therapies to improve synaptic plasticity are still limited and ineffectual, leading to sub-optimal recovery in patients. Therefore, promoting synaptic plasticity to ameliorate sensory-motor function may be a promising strategy for long-term recovery in stroke patients. Here, we review the involvement of different molecular pathways of glutamate and glia-mediated synapse loss, current pharmacological targets, and the emerging novel approaches to improve synaptic plasticity and sensory-motor impairment post-stroke.

Enhancing lower-limb rehabilitation: a scoping review of augmented reality environment

BACKGROUND

Lower-limb rehabilitation is crucial for restoring motor function in individuals with physical impairments; however, traditional rehabilitation approaches often encounter challenges such as limited resources and reduced patient motivation. Augmented reality (AR) offers an innovative approach by enriching rehabilitation with interactive and engaging experiences, thereby enhancing both motivation and treatment outcomes. AR environments enable patients to practice exercises in an immersive setting that emulates real-life scenarios, potentially increasing adherence and improving functional recovery.

METHODS

This scoping review analyzed 25 peer-reviewed studies on the use of AR within the "Environment" component of the Human-Computer-Environment system for lower-limb rehabilitation. We present a taxonomy of existing AR systems, categorizing them by rehabilitation tasks (content) and interaction modes (form), which identify both physical and virtual elements that contribute to a supportive AR environment.

DISCUSSION

The findings suggest that well-designed AR environments offer a flexible and cost-effective approach to various rehabilitation tasks. Customization is essential for addressing specific rehabilitation stages, including muscle strengthening, balance improvement, and gait training. The integration of multisensory feedback, such as visual, auditory, and haptic cues, enhances patient engagement and provides real-time performance monitoring. Effective AR environments must also account for the distinct needs of each limb, particularly for bilateral impairments, and ensure sufficient space for safe movement. By providing an individualized rehabilitation experience, AR environments have the potential to significantly improve patient motivation and outcomes. Future research should explore the integration of AR environments with assistive technologies, such as wearable devices and exoskeletons, to further enhance rehabilitation possibilities.

Large-scale fMRI dataset for the design of motor-based Brain-Computer Interfaces

Functional Magnetic Resonance Imaging (fMRI) data is commonly used to map sensorimotor cortical organization and to localise electrode target sites for implanted Brain-Computer Interfaces (BCIs). Functional data recorded during motor and somatosensory tasks from both adults and children specifically designed to map and localise BCI target areas throughout the lifespan is rare. Here, we describe a large-scale dataset collected from 155 human participants while they performed motor and somatosensory tasks involving the fingers, hands, arms, feet, legs, and mouth region. The dataset includes data from both adults and children (age range: 6-89 years) performing a set of standardized tasks. This dataset is particularly relevant to study developmental patterns in motor representation on the cortical surface and for the design of paediatric motor-based implanted BCIs.

Age-Related Changes in Action Observation EEG Response and Its Effect on BCI Performance

Action observation-based brain-computer interface (AO-BCI) can simultaneously elicit steady-state motion visual evoked potential in the occipital region and sensorimotor rhythm in the sensorimotor region, demonstrating substantial potential in neurorehabilitation applications. While current AO-BCI research primarily focuses on the younger population, this study conducted a comparative investigation of age-related differences in EEG response to the AO-BCI by enrolling 18 older and 18 younger subjects. We employed task discriminant component analysis (TDCA) to decode observed actions and performed comprehensive analyses of prefrontal EEG responses, i.e. approximate entropy (ApEn), sample entropy (SamEn), and rhythm power ratios (RPR), and the whole-brain functional network. Regression analyses were subsequently conducted to analyze the effects on the classification accuracy. Results revealed significantly diminished TDCA accuracy in older subjects (77.01% $\pm ~14.67$ %) compared to younger subjects (87.22% $\pm ~15.22$ %). Age-dependent EEG responses emerged across multiple dimensions: 1) Prefrontal ApEn, SamEn, and RPR exhibited distinct aging patterns; 2) Brain network analysis uncovered significant intergroup differences in $\alpha $ and $\beta $ band connectivity strength; 3) $\theta $ band network topology demonstrated reduced prefrontal nodal degree along with enhanced global efficiency in older subjects. Regression analysis identified a robust inverse relationship between the $\beta $ / $\theta $ RPR during stimulation and overall accuracy. And the $\beta $ / $\theta $ RPR and the $\beta $ band ApEn might be the main factor that causing individual differences in the identification accuracy in older and younger subjects, respectively. This study provides novel insights into age-related neuro-mechanisms in AO-BCI, establishing quantitative relationships between specific EEG features and BCI performance. These findings would offer guidelines for optimizing AO-BCI in rehabilitation.

Graph vertex and spectral features for EEG-based motor imagery classification

Motor imagery (MI) patterns play a vital role in brain-computer interface (BCI) systems, enabling control of external devices without relying on peripheral nerves or muscles. These patterns are typically classified by analyzing the associated electroencephalogram (EEG) signals. In this work, we introduce a novel MI classification approach based on multilevel graph-theoretic modeling of multichannel EEG signals. Multivariate autoregressive modeling and coherence analysis are firstly employed to construct directed graph signals to represent the relationships among EEG channels and capture the complex correlations inherent in MI patterns. Spatial graph vertex features are thus extracted as well as graph Fourier transform coefficients. Moreover, multilevel generalizations of vertex-domain features are thus defined where edges of graph signals are pruned according to different thresholds, vertex features are extracted for each threshold level, and then all features are combined into a multilevel hierarchical graph descriptor. These graph-theoretic descriptors could be fused with different variants of common spatial patterns for improved discriminability on MI classification tasks. Different feature combinations are used to train k-nearest neighbor classifiers, support vector machines, and random forests for MI pattern classification. The proposed method demonstrates competitive performance compared to the FWCSP and SCSP methods on Dataset 2a of the BCI Competition IV, as well as robust results on Dataset 1 from the same competition. Overall, the findings highlight the potential of multilevel spatial and spectral graph features in leveraging the correlation among EEG channels towards enhanced MI classification performance.

Material Selection and Device Design of Scalable Flexible Brain-Computer Interfaces: A Balance Between Electrical and Mechanical Performance

Brain-computer interfaces (BCIs) hold the potential to revolutionize brain function restoration, enhance human capability, and advance our understanding of cognitive mechanisms by directly linking neural signals with hardware. However, the mechanical mismatch between conventional rigid BCIs and soft brain tissue limits long-term interface stability. Next-generation BCIs must achieve long-term biocompatibility while maintaining high performance, enabling the integration of millions of sensors within tissue-level flexible and soft, stable neural interfaces. Lithographic fabrication techniques provide scalable thin-film flexible electronics, but traditional electronic materials often fail to meet the unique requirements of BCIs. This review examines the selection of materials and device design for flexible BCIs, starting with an analysis of intrinsic material properties-Young's modulus, electrical conductivity and dielectric constant. It then explores the integration of material selection with electrode design to optimize electrical circuits and assess key mechanical factors. Next, the correlation between electrical and mechanical performance is analyzed to guide material selection and device design. Finally, recent advances in neural probes are reviewed, highlighting improvements in signal quality, recording stability, and scalability. This review focuses on scalable, lithography-based BCIs, aiming to identify optimal materials and designs for long-term, reliable neural recordings.

Research on Adaptive Discriminating Method of Brain-Computer Interface for Motor Imagination

(1) Background: Brain-computer interface (BCI) technology represents a cutting-edge field that integrates brain intelligence with machine intelligence. Unlike BCIs that rely on external stimuli, motor imagery-based BCIs (MI-BCIs) generate usable brain signals based on an individual's imagination of specific motor actions. Due to the highly individualized nature of these signals, identifying individuals who are better suited for MI-BCI applications and improving its efficiency is critical. (2) Methods: This study collected four motor imagery tasks (left hand, right hand, foot, and tongue) from 50 healthy subjects and evaluated MI-BCI adaptability through classification accuracy. Functional networks were constructed using the weighted phase lag index (WPLI), and relevant graph theory parameters were calculated to explore the relationship between motor imagery adaptability and functional networks. (3) Results: Research has demonstrated a strong correlation between the network characteristics of tongue imagination and MI-BCI adaptability. Specifically, the nodal degree and characteristic path length in the right hemisphere were found to be significantly correlated with classification accuracy (p < 0.05). (4) Conclusions: The findings of this study offer new insights into the functional network mechanisms of motor imagery, suggesting that tongue imagination holds potential as a predictor of MI-BCI adaptability.

Simultaneous EEG and fNIRS recordings for semantic decoding of imagined animals and tools

Semantic neural decoding aims to identify which semantic concepts an individual focuses on at a given moment based on recordings of their brain activity. We investigated the feasibility of semantic neural decoding to develop a new type of brain-computer interface (BCI) that allows direct communication of semantic concepts, bypassing the character-by-character spelling used in current BCI systems. We provide data from our study to differentiate between two semantic categories of animals and tools during a silent naming task and three intuitive sensory-based imagery tasks using visual, auditory, and tactile perception. Participants were instructed to visualize an object (animal or tool) in their minds, imagine the sounds produced by the object, and imagine the feeling of touching the object. Simultaneous electroencephalography (EEG) and near-infrared spectroscopy (fNIRS) signals were recorded from 12 participants. Additionally, EEG signals were recorded from 7 other participants in a follow-up experiment focusing solely on the auditory imagery task. These datasets can serve as a valuable resource for researchers investigating semantic neural decoding, brain-computer interfaces, and mental imagery.

Comparing MEG and EEG measurement set-ups for a brain-computer interface based on selective auditory attention

Auditory attention modulates auditory evoked responses to target vs. non-target sounds in electro- and magnetoencephalographic (EEG/MEG) recordings. Employing whole-scalp MEG recordings and offline classification algorithms has been shown to enable high accuracy in tracking the target of auditory attention. Here, we investigated the decrease in accuracy when moving from the whole-scalp MEG to lower channel count EEG recordings and when training the classifier only from the initial or middle part of the recording instead of extracting training trials throughout the recording. To this end, we recorded simultaneous MEG (306 channels) and EEG (64 channels) in 18 healthy volunteers while presented with concurrent streams of spoken "Yes"/"No" words and instructed to attend to one of them. We then trained support vector machine classifiers to predict the target of attention from unaveraged trials of MEG/EEG. Classifiers were trained on 204 MEG gradiometers or on EEG with 64, 30, nine or three channels with trials extracted randomly across or only from the beginning of the recording. The highest classification accuracy, 73.2% on average across the participants for one-second trials, was obtained with MEG when the training trials were randomly extracted throughout the recording. With EEG, the accuracy was 69%, 69%, 66%, and 61% when using 64, 30, nine, and three channels, respectively. When training the classifiers with the same amount of data but extracted only from the beginning of the recording, the accuracy dropped by 11%-units on average, causing the result from the three-channel EEG to fall below the chance level. The combination of five consecutive trials partially compensated for this drop such that it was one to 5%-units. Although moving from whole-scalp MEG to EEG reduces classification accuracy, usable auditory-attention-based brain-computer interfaces can be implemented with a small set of optimally placed EEG channels.

Toward brain-computer interface speller with movement-related cortical potentials as control signals

Brain Computer Interface spellers offer a promising alternative for individuals with Amyotrophic Lateral Sclerosis (ALS) by facilitating communication without relying on muscle activity. This study assessed the feasibility of using movement related cortical potentials (MRCPs) as a control signal for a Brain-Computer Interface speller in an offline setting. Unlike motor imagery-based BCIs, this study focused on executed movements. Fifteen healthy subjects performed three spelling tasks that involved choosing specific letters displayed on a computer screen by performing a ballistic dorsiflexion of the dominant foot. Electroencephalographic signals were recorded from 10 sites centered around Cz. Three conditions were tested to evaluate MRCP performance under varying task demands: a control condition using repeated selections of the letter "O" to isolate movement-related brain activity; a phrase spelling condition with structured text ("HELLO IM FINE") to simulate a meaningful spelling task with moderate cognitive load; and a random condition using a randomized sequence of letters to introduce higher task complexity by removing linguistic or semantic context. The success rate, defined as the presence of an MRCP, was manually determined. It was approximately 69% for both the control and phrase conditions, with a slight decrease in the random condition, likely due to increased task complexity. Significant differences in MRCP features were observed between conditions with Laplacian filtering, whereas no significant differences were found in single-site Cz recordings. These results contribute to the development of MRCP-based BCI spellers by demonstrating their feasibility in a spelling task. However, further research is required to implement and validate real-time applications.

Multi-Scale Pyramid Squeeze Attention Similarity Optimization Classification Neural Network for ERP Detection

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

Comparing a BCI communication system in a patient with Multiple System Atrophy, with an animal model

Paralysis affects many people worldwide, and the people affected often suffer from impaired communication. We developed a microelectrode-based Brain-Computer Interface (BCI) for enabling communication in patients affected by paralysis, and implanted it in a patient with Multiple System Atrophy (MSA), a neurodegenerative disease that causes widespread neural symptoms including paralysis. To verify the effectiveness of the BCI system, it was also tested by implanting it in a non-human primate (NHP). Data from the human and NHP were used to train binary classifiers two different types of machine learning models: a Linear Discriminant Analysis (LDA) model, and a Long Short-Term Memory (LSTM)-based Artificial Neural Network (ANN). The LDA model performed at up to 72.7 % accuracy for binary decoding in the human patient, however, performance was highly variable and was much lower on most recording days. The BCI system was able to accurately decode movement vs non-movement in the NHP (accuracy using LDA: 82.7 ± 3.3 %, LSTM: 83.7 ± 2.2 %, 95 % confidence intervals), however it was not able to with recordings from the human patient (accuracy using LDA: 47.0 ± 5.1 %, LSTM: 44.6 ± 9.9 %, 95 % confidence intervals). We discuss how neurodegenerative diseases such as MSA can impede BCI-based communication, and postulate on the mechanisms by which this may occur.

Fabrication of micro-wire stent electrode as a minimally invasive endovascular neural interface for vascular electrocorticography using laser ablation method

Objective. Minimally invasive endovascular stent electrode is a currently emerging technology in neural engineering with minimal damage to the neural tissue. However, the typical stent electrode still requires resistive welding and is relatively large, limiting its application mainly on the large animal or thick vessels. In this study, we investigated the feasibility of laser ablation of micro-wire stent electrode with a diameter as small as 25μmand verified it in the superior sagittal sinus (SSS) of a rat.Approach. We have developed a laser ablation technology to expose the electrode sites of the micro-wire on both sides without damaging the wire itself. During laser ablation, we applied a new method to fix and realign the micro-wires. The micro-wire stent electrode was fabricated by carefully assemble the micro-wire and stent. We tested the electrochemical performances of the electrodes as a neural interface. Finally, we deployed the stent electrode in a rat to verified the feasibility.Main result. Based on the proposed micro-wire stent electrode, we demonstrated that the stent electrode could be successfully deployed in a rat. With the benefit of the smaller design and laser fabrication technology, it can be fitted into a catheter with an inner diameter of 0.6mm. The vascular electrocorticography can be detected during the acute recording, making it promising in the application of small animals and thin vessels.Significance. The method we proposed combines the advantages of endovascular micro-wire electrode and stent, helping make the electrodes smaller. This study provided an alternative method for deploying micro-wire electrodes into thinner vessels as an endovascular neural interface.

A wearable brain-computer interface to play an endless runner game by self-paced motor imagery

Objective.A wearable brain-computer interface is proposed and validated experimentally in relation to the real-time control of an endless runner game by self-paced motor imagery(MI).Approach.Electroencephalographic signals were recorded via eight wet electrodes. The processing pipeline involved a filter-bank common spatial pattern approach and the combination of three binary classifiers exploiting linear discriminant analysis. This enabled the discrimination between imagining left-hand, right-hand, and no movement. Each mental task corresponded to an avatar horizontal motion within the game. Twenty-three healthy subjects participated to the experiments and their data are made publicly available. A custom metric was proposed to assess avatar control performance during the gaming phase. The game consisted of two levels, and after each, participants completed a questionnaire to self-assess their engagement and gaming experience.Main results.The mean classification accuracies resulted 73%, 73%, and 67% for left-rest, right-rest, and left-right discrimination, respectively. In the gaming phase, subjects with higher accuracies for left-rest and right-rest pair exhibited higher performance in terms of the custom metric. Correlation of the offline and real-time performance was investigated. The left-right MI did not correlate to the gaming phase performance due to the poor mean accuracy of the calibration. Finally, the engagement questionnaires revealed that level 1 and level 2 were not perceived as frustrating, despite the increasing difficulty.Significance.The work contributes to the development of wearable and self-paced interfaces for real-time control. These enhance user experience by guaranteeing a more natural interaction with respect to synchronous neural interfaces. Moving beyond benchmark datasets, the work paves the way to future applications on mobile devices for everyday use.

Single-Trial Electroencephalography Discrimination of Real, Regulated, Isometric Wrist Extension and Wrist Flexion

Improved interpretation of electroencephalography (EEG) associated with the neural control of essential hand movements, including wrist extension (WE) and wrist flexion (WF), could improve the performance of brain-computer interfaces (BCIs). These BCIs could control a prosthetic or orthotic hand to enable motor-impaired individuals to regain the performance of activities of daily living. This study investigated the interpretation of neural signal patterns associated with kinematic differences between real, regulated, isometric WE and WF movements from recorded EEG data. We used 128-channel EEG data recorded from 14 participants performing repetitions of the wrist movements, where the force, speed, and range of motion were regulated. The data were filtered into four frequency bands: delta and theta, mu and beta, low gamma, and high gamma. Within each frequency band, independent component analysis was used to isolate signals originating from seven cortical regions of interest. Features were extracted from these signals using a time-frequency algorithm and classified using Mahalanobis distance clustering. We successfully classified bilateral and unilateral WE and WF movements, with respective accuracies of 90.68% and 69.80%. The results also demonstrated that all frequency bands and regions of interest contained motor-related discriminatory information. Bilateral discrimination relied more on the mu and beta bands, while unilateral discrimination favoured the gamma bands. These results suggest that EEG-based BCIs could benefit from the extraction of features from multiple frequencies and cortical regions.

Simulating online and offline tasks using hybrid cheetah optimization algorithm for patients affected by neurodegenerative diseases

Brain-Computer Interface (BCI) is a versatile technique to offer better communication system for people affected by the locked-in syndrome (LIS).In the current decade, there has been a growing demand for improved care and services for individuals with neurodegenerative diseases. To address this barrier, the current work is designed with four states of BCI for paralyzed persons using Welch Power Spectral Density (W-PSD). The features extracted from the signals were trained with a hybrid Feed Forward Neural Network Cheetah Optimization Algorithm (FFNNCOA) in both offline and online modes. Totally, eighteen subjects were involved in this study. The study proved that the offline analysis phase outperformed than the online phase in the real-time. The experiment was achieved the accuracies of 95.56% and 93.88% for men and female respectively. Furthermore, the study confirms that the subject's performance in the offline can manage the task more easily than in online mode.

HIPPIE: A Multimodal Deep Learning Model for Electrophysiological Classification of Neurons

Extracellular electrophysiological recordings present unique computational challenges for neuronal classification due to noise, technical variability, and batch effects across experimental systems. We introduce HIPPIE (High-dimensional Interpretation of Physiological Patterns In Extracellular recordings), a deep learning framework that combines self-supervised pretraining on unlabeled datasets with supervised fine-tuning to classify neurons from extracellular recordings. Using conditional convolutional joint autoencoders, HIPPIE learns robust, technology-adjusted representations of waveforms and spiking dynamics. This model can be applied to electrophysiological classification and clustering across diverse biological cultures and technologies. We validated HIPPIE on both in vivo mouse recordings and in vitro brain slices, where it demonstrated superior performance over other unsupervised methods in cell-type discrimination and aligned closely with anatomically defined classes. Its latent space organizes neurons along electrophysiological gradients, while enabling batch and individual corrected alignment of recordings across experiments. HIPPIE establishes a general framework for systematically decoding neuronal diversity in native and engineered systems.

Global trends and hotspots of neuromodulation in spinal cord injury: a study based on bibliometric analysis

OBJECTIVE

Spinal cord injury (SCI) is a debilitating condition that can result in permanent disability. Neuromodulation is a promising technology that has gained popularity as a treatment for SCI. This study aims to analyze the published literature to investigate the global trends and hotspots in research on neuromodulation in the context of SCI.

METHODS

All relevant publications on the topic of neuromodulation in SCI from January 1, 2005, to September 17, 2024, were acquired from the Web of Science Core Collection database. Bibliometric analysis was performed to evaluate the publication distribution by country, institution, author, and journal, as well as keyword, using CiteSpace, VOSviewer, and Scimago Graphica software.

RESULTS

Overall, 3,211 publications were eligible for inclusion in the analysis. The publication number in 2005 and 2024 were 77 and 222, respectively. A steady increasing trend in the publication number over the past two decades was observed. The Unites States published 1544 articles with 52,521 citations, ranking first regarding publication number and total citations. Case Western Reserve University was the most productive institution that published 181 papers. All of the highly productive institutions were located in the United States, Canada, and Australia. The University of California Los Angeles harvested 6626 total citations and 81.8 average citations, ranking first among the productive institutions. Gorgey AS published 60 articles and ranked first regarding total publication number. Edgerton VR harvested 4333 citations and ranked first among the authors for total citations. The analysis of high-yielding journals suggested that Journal of Spinal Cord Medicine was the most productive journal with 133 publications. Spinal Cord yielded 4200 citations and ranked first among the journals for total citations. The keyword analysis identified "functional electrical stimulation" and "spinal cord stimulation" as research hotspots.

CONCLUSION

This study delineates the current knowledge landscape and research trends on the topic of neuromodulation in SCI. The findings highlight the growing interest in this field and underscore the significance of neuromodulation in SCI research.

The state-of-the-art of invasive brain-computer interfaces in humans: a systematic review and individual patient meta-analysis

Objective.Invasive brain-computer interfaces (iBCIs) have evolved significantly since the first neurotrophic electrode was implanted in a human subject three decades ago. Since then, both hardware and software advances have increased the iBCI performance to enable tasks such as decoding conversations in real-time and manipulating external limb prostheses with haptic feedback. In this systematic review, we aim to evaluate the advances in iBCI hardware, software and functionality and describe challenges and opportunities in the iBCI field.Approach.Medline, EMBASE, PubMed and Cochrane databases were searched from inception until 13 April 2024. Primary studies reporting the use of iBCI in human subjects to restore function were included. Endpoints extracted include iBCI electrode type, iBCI implantation, decoder algorithm, iBCI effector, testing and training methodology and functional outcomes. Narrative synthesis of outcomes was done with a focus on hardware and software development trends over time. Individual patient data (IPD) was also collected and an IPD meta-analysis was done to identify factors significant to iBCI performance.Main results.93 studies involving 214 patients were included in this systematic review. The median task performance accuracy for cursor control tasks was 76.00% (Interquartile range [IQR] = 21.2), for motor tasks was 80.00% (IQR = 23.3), and for communication tasks was 93.27% (IQR = 15.3). Current advances in iBCI software include use of recurrent neural network architectures as decoders, while hardware advances such as intravascular stentrodes provide a less invasive alternative for neural recording. Challenges include the lack of standardized testing paradigms for specific functional outcomes and issues with portability and chronicity limiting iBCI usage to laboratory settings.Significance.Our systematic review demonstrated the exponential rate at which iBCIs have evolved over the past two decades. Yet, more work is needed for widespread clinical adoption and translation to long-term home-use.

Signal Characteristics, Motor Cortex Engagement, and Classification Performance of Combined Action Observation, Motor Imagery and SSMVEP (CAMS) BCI

Motor imagery (MI)-based Brain-Computer Interfaces (BCIs) have shown promise in engaging the motor cortex for recovery. However, individual responses to MI-based BCIs are highly variable and relatively weak. Conversely, combined action observation (AO) and motor imagery (MI) paradigms have demonstrated stronger responses compared to AO or MI alone, along with enhanced cortical excitability. In this study, a novel BCI called Combined AO, MI, and Steady-State Motion Visual Evoked Potential (SSMVEP) (CAMS) was proposed. CAMS was designed based on gait observation and imagination. Twenty-five healthy volunteers participated in the study with CAMS serving as the intervention and SSMVEP checkerboard as the control condition. We hypothesized the CAMS intervention can induce observable increases in the negativity of the movement-related cortical potential (MRCP) associated with ankle dorsiflexion. MRCP components, including Bereitschaftspotential, were measured pre- and post-intervention. Additionally, the signal characteristics of the visual and motor responses were quantified. Finally, a two-class visual BCI classification performance was assessed. A consistent increase in negativity was observed across all MRCP components in signals over the primary motor cortex, compared to the control condition. CAMS visual BCI achieved a median accuracy of 83.8%. These findings demonstrate the ability of CAMS BCI to enhance cortical excitability in relation to movement preparation and execution. The CAMS stimulus not only evokes SSMVEP-like activity and sensorimotor rhythm but also enhances the MRCP. These findings contribute to the understanding of CAMS paradigm in enhancing cortical excitability, consistent and reliable classification performance holding promise for motor rehabilitation outcomes and future BCI design considerations.

Decoding thoughts, encoding ethics: A narrative review of the BCI-AI revolution

OBJECTIVES

This narrative review aims to analyze mechanisms underlying Brain-Computer Interface (BCI) and Artificial Intelligence (AI) integration, evaluate recent advances in signal acquisition and processing techniques, and assess AI-enhanced neural decoding strategies. The review identifies critical research gaps and examines emerging solutions across multiple domains of BCI-AI integration.

METHODS

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

RESULTS

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

CONCLUSIONS

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

Stimulus-Informed Generalized Canonical Correlation Analysis for Group Analysis of Neural Responses to Natural Stimuli

Various new brain-computer interface technologies or neuroscience applications require decoding stimulus-following neural responses to natural stimuli such as speech and video from, e.g., electroencephalography (EEG) signals. In this context, generalized canonical correlation analysis (GCCA) is often used as a group analysis technique, which allows the extraction of correlated signal components from the neural activity of multiple subjects attending to the same stimulus. GCCA can be used to improve the signal-to-noise ratio of the stimulus-following neural responses relative to all other irrelevant (non-)neural activity, or to quantify the correlated neural activity across multiple subjects in a group-wise coherence metric. However, the traditional GCCA technique is stimulus-unaware: no information about the stimulus is used to estimate the correlated components from the neural data of several subjects. Therefore, the GCCA technique might fail to extract relevant correlated signal components in practical situations where the amount of information is limited, for example, because of a limited amount of training data or group size. This motivates a new stimulus-informed GCCA (SI-GCCA) framework that allows taking the stimulus into account to extract the correlated components. We show that SI-GCCA outperforms GCCA in various practical settings, for both auditory and visual stimuli. Moreover, we showcase how SI-GCCA can be used to steer the estimation of the components towards the stimulus. As such, SI-GCCA substantially improves upon GCCA for various purposes, ranging from preprocessing to quantifying attention.

Gamma oscillation optimally predicts finger movements

Our fingers are the most dexterous and complicated parts of our body and play a significant role in our daily activities. Non-invasive techniques, such as Electroencephalography (EEG) and Electromyography (EMG) can be used to collect neural and muscular signals related to finger movements. In this study, we combined an 8-channel EMG and a 31-channel EEG while the human subject moved one of the five fingers on the right hand. To identify the best EEG frequency features that encode distinct finger movements, we systematically examined the decoding accuracies of the slow-cortical potentials and three types of sensorimotor rhythms, namely the Mu, beta, and gamma oscillations. For both EMG and EEG, we came up with a simple and unified root mean square or power approach that avoided the complex signal features used by previous studies. The signal features were then fed into a feedforward artificial-neural-network (ANN) classifier. We found that the low-gamma oscillation provided the best decoding performance over the other frequency bands, ranging from 65.0 % to 89.0 %, which was comparable to the EMG performance. Combining EMG and low gamma into a single ANN can further improve the outcome for subjects who had showed suboptimal performances with EMG or EEG alone. This study provided a simple and efficient algorithm for prosthetics that assist patients with sensorimotor impairments.

onEEGwaveLAD: A fully automated online EEG wavelet-based learning adaptive denoiser for artefacts identification and mitigation

Electroencephalographic signals are obtained by amplifying and recording the brain's spontaneous biological potential using electrodes positioned on the scalp. While proven to help find changes in brain activity with a high temporal resolution, such signals are contaminated by non-stationary and frequent artefacts. A plethora of noise reduction techniques have been developed, achieving remarkable performance. However, they often require multi-channel information and additional reference signals, are not fully automated, require human intervention and are mostly offline. With the popularity of Brain-Computer Interfaces and the application of Electroencephalography in daily activities and other ecological settings, there is an increasing need for robust, online, near real-time denoising techniques, without additional reference signals, that is fully automated and does not require human supervision nor multi-channel information. This research contributes to the body of knowledge by introducing onEEGwaveLAD, a novel, fully automated, ONline, EEG wavelet-based Learning Adaptive Denoiser pipeline for artefact identification and reduction. It is a specific framework that can be instantiated for various types of artefacts paving the path towards real-time denoising. As the first of its kind, it is described and instantiated for the particular problem of blink detection and reduction, and evaluated across a general and a specific analysis of the signal to noise ratio across 30 participants.

Wearable Biodevices Based on Two-Dimensional Materials: From Flexible Sensors to Smart Integrated Systems

The proliferation of wearable biodevices has boosted the development of soft, innovative, and multifunctional materials for human health monitoring. The integration of wearable sensors with intelligent systems is an overwhelming tendency, providing powerful tools for remote health monitoring and personal health management. Among many candidates, two-dimensional (2D) materials stand out due to several exotic mechanical, electrical, optical, and chemical properties that can be efficiently integrated into atomic-thin films. While previous reviews on 2D materials for biodevices primarily focus on conventional configurations and materials like graphene, the rapid development of new 2D materials with exotic properties has opened up novel applications, particularly in smart interaction and integrated functionalities. This review aims to consolidate recent progress, highlight the unique advantages of 2D materials, and guide future research by discussing existing challenges and opportunities in applying 2D materials for smart wearable biodevices. We begin with an in-depth analysis of the advantages, sensing mechanisms, and potential applications of 2D materials in wearable biodevice fabrication. Following this, we systematically discuss state-of-the-art biodevices based on 2D materials for monitoring various physiological signals within the human body. Special attention is given to showcasing the integration of multi-functionality in 2D smart devices, mainly including self-power supply, integrated diagnosis/treatment, and human-machine interaction. Finally, the review concludes with a concise summary of existing challenges and prospective solutions concerning the utilization of 2D materials for advanced biodevices.

Support matrix machine: A review

Support vector machine (SVM) is one of the most studied paradigms in the realm of machine learning for classification and regression problems. It relies on vectorized input data. However, a significant portion of the real-world data exists in matrix format, which is given as input to SVM by reshaping the matrices into vectors. The process of reshaping disrupts the spatial correlations inherent in the matrix data. Also, converting matrices into vectors results in input data with a high dimensionality, which introduces significant computational complexity. To overcome these issues in classifying matrix input data, support matrix machine (SMM) is proposed. It represents one of the emerging methodologies tailored for handling matrix input data. SMM preserves the structural information of the matrix data by using the spectral elastic net property which is a combination of the nuclear norm and Frobenius norm. This article provides the first in-depth analysis of the development of the SMM model, which can be used as a thorough summary by both novices and experts. We discuss numerous SMM variants, such as robust, sparse, class-imbalance, and multi-class classification models. We also analyze the applications of the SMM and conclude the article by outlining potential future research avenues and possibilities that may motivate researchers to advance the SMM algorithm.

A temporal-spectral fusion transformer with subject-specific adapter for enhancing RSVP-BCI decoding

The Rapid Serial Visual Presentation (RSVP)-based Brain-Computer Interface (BCI) is an efficient technology for target retrieval using electroencephalography (EEG) signals. The performance improvement of traditional decoding methods relies on a substantial amount of training data from new test subjects, which increases preparation time for BCI systems. Several studies introduce data from existing subjects to reduce the dependence of performance improvement on data from new subjects, but their optimization strategy based on adversarial learning with extensive data increases training time during the preparation procedure. Moreover, most previous methods only focus on the single-view information of EEG signals, but ignore the information from other views which may further improve performance. To enhance decoding performance while reducing preparation time, we propose a Temporal-Spectral fusion transformer with Subject-specific Adapter (TSformer-SA). Specifically, a cross-view interaction module is proposed to facilitate information transfer and extract common representations across two-view features extracted from EEG temporal signals and spectrogram images. Then, an attention-based fusion module fuses the features of two views to obtain comprehensive discriminative features for classification. Furthermore, a multi-view consistency loss is proposed to maximize the feature similarity between two views of the same EEG signal. Finally, we propose a subject-specific adapter to rapidly transfer the knowledge of the model trained on data from existing subjects to decode data from new subjects. Experimental results show that TSformer-SA significantly outperforms comparison methods and achieves outstanding performance with limited training data from new subjects. This facilitates efficient decoding and rapid deployment of BCI systems in practical use.

Signal acquisition of brain-computer interfaces: A medical-engineering crossover perspective review

Brain-computer interface (BCI) technology represents a burgeoning interdisciplinary domain that facilitates direct communication between individuals and external devices. The efficacy of BCI systems is largely contingent upon the progress in signal acquisition methodologies. This paper endeavors to provide an exhaustive synopsis of signal acquisition technologies within the realm of BCI by scrutinizing research publications from the last ten years. Our review synthesizes insights from both clinical and engineering viewpoints, delineating a comprehensive two-dimensional framework for understanding signal acquisition in BCIs. We delineate nine discrete categories of technologies, furnishing exemplars for each and delineating the salient challenges pertinent to these modalities. This review furnishes researchers and practitioners with a broad-spectrum comprehension of the signal acquisition landscape in BCI, and deliberates on the paramount issues presently confronting the field. Prospective enhancements in BCI signal acquisition should focus on harmonizing a multitude of disciplinary perspectives. Achieving equilibrium between signal fidelity, invasiveness, biocompatibility, and other pivotal considerations is imperative. By doing so, we can propel BCI technology forward, bolstering its effectiveness, safety, and dependability, thereby contributing to an auspicious future for human-technology integration.

Laser Welding of Micro-Wire Stent Electrode as a Minimally Invasive Endovascular Neural Interface

Minimally invasive endovascular stent electrodes are an emerging technology in neural engineering, designed to minimize the damage to neural tissue. However, conventional stent electrodes often rely on resistive welding and are relatively bulky, restricting their use primarily to large animals or thick blood vessels. In this study, the feasibility is explored of fabricating a laser welding stent electrode as small as 300 μm. A high-precision laser welding technique was developed to join micro-wire electrodes without compromising structural integrity or performance. To ensure consistent results, a novel micro-wire welding with platinum pad method was introduced during the welding process. The fabricated electrodes were integrated with stent structures and subjected to detailed electrochemical performance testing to evaluate their potential as neural interface components. The laser-welded endovascular stent electrodes exhibited excellent electrochemical properties, including low impedance and stable charge transfer capabilities. At the same time, in this study, a simulation is conducted of the electrode distribution and arrangement on the stent structure, optimizing the utilization of the available surface area for enhanced functionality. These results demonstrate the potential of the fabricated electrodes for high-performance neural interfacing in endovascular applications. The approach provided a promising solution for advancing endovascular neural engineering technologies, particularly in applications requiring compact electrode designs.

The Effect of Processing Techniques on the Classification Accuracy of Brain-Computer Interface Systems

Background/Objectives: Accurately classifying Electroencephalography (EEG) signals is essential for the effective operation of Brain-Computer Interfaces (BCI), which is needed for reliable neurorehabilitation applications. However, many factors in the processing pipeline can influence classification performance. The objective of this study is to assess the effects of different processing steps on classification accuracy in EEG-based BCI systems. Methods: This study explores the impact of various processing techniques and stages, including the FASTER algorithm for artifact rejection (AR), frequency filtering, transfer learning, and cropped training. The Physionet dataset, consisting of four motor imagery classes, was used as input due to its relatively large number of subjects. The raw EEG was tested with EEGNet and Shallow ConvNet. To examine the impact of adding a spatial dimension to the input data, we also used the Multi-branch Conv3D Net and developed two new models, Conv2D Net and Conv3D Net. Results: Our analysis showed that classification accuracy can be affected by many factors at every stage. Applying the AR method, for instance, can either enhance or degrade classification performance, depending on the subject and the specific network architecture. Transfer learning was effective in improving the performance of all networks for both raw and artifact-rejected data. However, the improvement in classification accuracy for artifact-rejected data was less pronounced compared to unfiltered data, resulting in reduced precision. For instance, the best classifier achieved 46.1% accuracy on unfiltered data, which increased to 63.5% with transfer learning. In the filtered case, accuracy rose from 45.5% to only 55.9% when transfer learning was applied. An unexpected outcome regarding frequency filtering was observed: networks demonstrated better classification performance when focusing on lower-frequency components. Higher frequency ranges were more discriminative for EEGNet and Shallow ConvNet, but only when cropped training was applied. Conclusions: The findings of this study highlight the complex interaction between processing techniques and neural network performance, emphasizing the necessity for customized processing approaches tailored to specific subjects and network architectures.

EEG channel and feature investigation in binary and multiple motor imagery task predictions

Introduction

Motor Imagery (MI) Electroencephalography (EEG) signals are non-stationary and dynamic physiological signals which have low signal-to-noise ratio. Hence, it is difficult to achieve high classification accuracy. Although various machine learning methods have already proven useful to that effect, the use of many features and ineffective EEG channels often leads to a complex structure of classifier algorithms. State-of-the-art studies were interested in improving classification performance with complex feature extraction and classification methods by neglecting detailed EEG channel and feature investigation in predicting MI tasks from EEGs. Here, we investigate the effects of the statistically significant feature selection method on four different feature domains (time-domain, frequency-domain, time-frequency domain, and non-linear domain) and their two different combinations to reduce the number of features and classify MI-EEG features by comparing low-dimensional matrices with well-known machine learning algorithms.

Methods

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

Results

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

Discussion

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

Olfactory stimulation in newborns: Regional differences in cerebral oxygenation

BACKGROUND

The sense of smell is fully developed in newborns and plays an important role in their early development. There are several approaches to studying olfactory processing in the newborn brain, including EEG, fMRI, and near-infrared spectroscopy (NIRS). Understanding the processing of olfactory stimuli in the newborn brain is of fundamental importance for the development of supportive therapeutic odorant delivery, e.g. for weaning by gavage, and for adapting it to the developing brain. This study aimed to investigate the effect of different odors (milk, farnesol odor, and water as a control) on changes in brain activation in newborns in two different brain regions.

METHODS

Newborns older than 72 h and below an age of seven days were divided into two groups with different optode positioning strategies of NIRS, group I parietal and group II frontal. Olfactory stimulation was administered using milk, farnesol (floral odor), and water as a control.

RESULTS

A total of 26 newborns participated in the study. In the final analysis, 19 children were included. Allthough the optode positioning does not differ significantly, in group I, farnesol stimulation resulted in a significant increase in oxygenated hemoglobin compared to the control, while milk odor showed a decreased amplitude, particularly in the more parietal optode position. In group II, a significant difference was observed between the milk odor and the control, in the frontal areas.

CONCLUSIONS

This study revealed significant changes in hemoglobin oxygenation, indicating neuronal activation following different olfactory stimulation in both optode positionings. Whereas milk had more impact in frontal areas, the floral odor caused an effect in parietal areas.

Media Representation of the Ethical Issues Pertaining to Brain-Computer Interface (BCI) Technology

BACKGROUND/OBJECTIVES

Brain-computer interfaces (BCIs) are a rapidly developing technology that captures and transmits brain signals to external sources, allowing the user control of devices such as prosthetics. BCI technology offers the potential to restore physical capabilities in the body and change how we interact and communicate with computers and each other. While BCI technology has existed for decades, recent developments have caused the technology to generate a host of ethical issues and discussions in both academic and public circles. Given that media representation has the potential to shape public perception and policy, it is necessary to evaluate the space that these issues take in public discourse.

METHODS

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

RESULTS

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

CONCLUSIONS

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

Brain-Computer Interface for Patients with Spinal Cord Injury: A Bibliometric Study

BACKGROUND

Spinal cord injury (SCI) is a debilitating condition with profound implications on patients' quality of life. Recent advancements in brain-computer interface (BCI) technology have provided novel opportunities for individuals with paralysis due to SCI. Consequently, research on the application of BCI for treating SCI has received increasing attention from scholars worldwide. However, there is a lack of rigorous bibliometric studies on the evolution and trends in this field. Hence, the present study aimed to use bibliometric methods to investigate the current status and emerging trends in the field of applying BCI for treating SCI and thus identify novel therapeutic options for SCI.

METHODS

We conducted a comprehensive review of the relevant literature on BCI applications for treating SCI published between 2005 and 2024 by using the Web of Science Core Collection database. To facilitate visualization and quantitative analysis of the published literature, we used VOSviewer and CiteSpace software tools. These tools enabled the assessment of co-authorships, co-occurrences, citations, and co-citations in the selected literature, thereby providing an overview of the current trends and predictive insights into the field.

RESULTS

The literature search yielded 714 publications from the Web of Science Core Collection database. The findings indicated a significant upward trend in the number of publications, yielding a total of 24,804 citations, with an average citation rate of 34.74 per publication and an H-index of 75. Research contributions were identified from 54 countries/regions, and the United States, China, and Germany emerged as the predominant contributors. A total of 1114 research institutions contributed to the retrieved literature, with Harvard Medical School, Brown University, and Northwestern University producing the highest number of publications. The published literature was predominantly distributed across 258 academic journals, and the Journal of Neural Engineering was the most frequently utilized publication source. Hochberg, Leigh, Henderson, Jaimie, and Collinger were the prominent authors in this field.

CONCLUSIONS

In recent years, there has been a steep increase in research on the use of BCI for treating SCI. Existing research focuses on the application of BCI for improving rehabilitation and quality of life of patients with SCI. Interdisciplinary collaboration is the current trend in this field.

Decoding lower-limb kinematic parameters during pedaling tasks using deep learning approaches and EEG

Stroke is a neurological condition that usually results in the loss of voluntary control of body movements, making it difficult for individuals to perform activities of daily living (ADLs). Brain-computer interfaces (BCIs) integrated into robotic systems, such as motorized mini exercise bikes (MMEBs), have been demonstrated to be suitable for restoring gait-related functions. However, kinematic estimation of continuous motion in BCI systems based on electroencephalography (EEG) remains a challenge for the scientific community. This study proposes a comparative analysis to evaluate two artificial neural network (ANN)-based decoders to estimate three lower-limb kinematic parameters: x- and y-axis position of the ankle and knee joint angle during pedaling tasks. Long short-term memory (LSTM) was used as a recurrent neural network (RNN), which reached Pearson correlation coefficient (PCC) scores close to 0.58 by reconstructing kinematic parameters from the EEG features on the delta band using a time window of 250 ms. These estimates were evaluated through kinematic variance analysis, where our proposed algorithm showed promising results for identifying pedaling and rest periods, which could increase the usability of classification tasks. Additionally, negative linear correlations were found between pedaling speed and decoder performance, thereby indicating that kinematic parameters between slower speeds may be easier to estimate. The results allow concluding that the use of deep learning (DL)-based methods is feasible for the estimation of lower-limb kinematic parameters during pedaling tasks using EEG signals. This study opens new possibilities for implementing controllers most robust for MMEBs and BCIs based on continuous decoding, which may allow for maximizing the degrees of freedom and personalized rehabilitation.

Decoding motor imagery loaded on steady-state somatosensory evoked potential based on complex task-related component analysis

BACKGROUND AND OBJECTIVE

Motor Imagery (MI) recognition is one of the most critical decoding problems in brain- computer interface field. Combined with the steady-state somatosensory evoked potential (MI-SSSEP), this new paradigm can achieve higher recognition accuracy than the traditional MI paradigm. Typical algorithms do not fully consider the characteristics of MI-SSSEP signals. Developing an algorithm that fully captures the paradigm's characteristics to reduce false triggering rate is the new step in improving performance.

METHODS

The idea to use complex signal task-related component analysis (cTRCA) algorithm for spatial filtering processing has been proposed in this paper according to the features of SSSEP signal. In this research, it's proved from the analysis of simulation signals that task-related component analysis (TRCA) as typical method is affected when the response between stimuli has reduced correlation and the proposed algorithm can effectively overcome this problem. The experimental data under the MI-SSSEP paradigm have been used to identify right-handed target tasks and three unique interference tasks are used to test the false triggering rate. cTRCA demonstrates superior performance as confirmed by the Wilcoxon signed-rank test.

RESULTS

The recognition algorithm of cTRCA combined with mutual information-based best individual feature (MIBIF) and minimum distance to mean (MDM) can obtain AUC value up to 0.89, which is much higher than traditional algorithm common spatial pattern (CSP) combined with support vector machine (SVM) (the average AUC value is 0.77, p < 0.05). Compared to CSP+SVM, this algorithm model reduced the false triggering rate from 38.69 % to 20.74 % (p < 0.001).

CONCLUSIONS

The research prove that TRCA is influenced by MI-SSSEP signals. The results further prove that the motor imagery task in the new paradigm MI-SSSEP causes the phase change in evoked potential. and the cTRCA algorithm based on such phase change is more suitable for this hybrid paradigm and more conducive to decoding the motor imagery task and reducing false triggering rate.

FDCN-C: A deep learning model based on frequency enhancement, deformable convolution network, and crop module for electroencephalography motor imagery classification

Motor imagery (MI)-electroencephalography (EEG) decoding plays an important role in brain-computer interface (BCI), which enables motor-disabled patients to communicate with external world via manipulating smart equipment. Currently, deep learning (DL)-based methods are popular for EEG decoding. Whereas the utilization efficiency of EEG features in frequency and temporal domain is not sufficient, which results in poor MI classification performance. To address this issue, an EEG-based MI classification model based on a frequency enhancement module, a deformable convolutional network, and a crop module (FDCN-C) is proposed. Firstly, the frequency enhancement module is innovatively designed to address the issue of extracting frequency information. It utilizes convolution kernels at continuous time scales to extract features across different frequency bands. These features are screened by calculating attention and integrated into the original EEG data. Secondly, for temporal feature extraction, a deformable convolution network is employed to enhance feature extraction capabilities, utilizing offset parameters to modulate the convolution kernel size. In spatial domain, a one-dimensional convolution layer is designed to integrate all channel information. Finally, a dilated convolution is used to form a crop classification module, wherein the diverse receptive fields of the EEG data are computed multiple times. Two public datasets are employed to verify the proposed FDCN-C model, the classification accuracy obtained from the proposed model is greater than that of state-of-the-art methods. The model's accuracy has improved by 14.01% compared to the baseline model, and the ablation study has confirmed the effectiveness of each module in the model.

Using data from cue presentations results in grossly overestimating semantic BCI performance

Neuroimaging studies have reported the possibility of semantic neural decoding to identify specific semantic concepts from neural activity. This offers promise for brain-computer interfaces (BCIs) for communication. However, translating these findings into a BCI paradigm has proven challenging. Existing EEG-based semantic decoding studies often rely on neural activity recorded when a cue is present, raising concerns about decoding reliability. To address this, we investigate the effects of cue presentation on EEG-based semantic decoding. In an experiment with a clear separation between cue presentation and mental task periods, we attempt to differentiate between semantic categories of animals and tools in four mental tasks. By using state-of-the-art decoding analyses, we demonstrate significant mean classification accuracies up to 71.3% during cue presentation but not during mental tasks, even with adapted analyses from previous studies. These findings highlight a potential issue when using neural activity recorded during cue presentation periods for semantic decoding. Additionally, our results show that semantic decoding without external cues may be more challenging than current state-of-the-art research suggests. By bringing attention to these issues, we aim to stimulate discussion and drive advancements in the field toward more effective semantic BCI applications.

IMH-Net: a convolutional neural network for end-to-end EEG motor imagery classification

As the main component of Brain-computer interface (BCI) technology, the classification algorithm based on EEG has developed rapidly. The previous algorithms were often based on subject-dependent settings, resulting in BCI needing to be calibrated for new users. In this work, we propose IMH-Net, an end-to-end subject-independent model. The model first uses Inception blocks extracts the frequency domain features of the data, then further compresses the feature vectors to extract the spatial domain features, and finally learns the global information and classification through Multi-Head Attention mechanism. On the OpenBMI dataset, IMH-Net obtained 73.90 ± 13.10% accuracy and 73.09 ± 14.99% F1-score in subject-independent manner, which improved the accuracy by 1.96% compared with the comparison model. On the BCI competition IV dataset 2a, this model also achieved the highest accuracy and F1-score in subject-dependent manner. The IMH-Net model we proposed can improve the accuracy of subject-independent Motor Imagery (MI), and the robustness of the algorithm is high, which has strong practical value in the field of BCI.

Adversarial artifact detection in EEG-based brain-computer interfaces

Objective. machine learning has achieved significant success in electroencephalogram (EEG) based brain-computer interfaces (BCIs), with most existing research focusing on improving the decoding accuracy. However, recent studies have shown that EEG-based BCIs are vulnerable to adversarial attacks, where small perturbations added to the input can cause misclassification. Detecting adversarial examples is crucial for both understanding this phenomenon and developing effective defense strategies.Approach. this paper, for the first time, explores adversarial detection in EEG-based BCIs. We extend several popular adversarial detection approaches from computer vision to BCIs. Two new Mahalanobis distance based adversarial detection approaches, and three cosine distance based adversarial detection approaches, are also proposed, which showed promising performance in detecting three kinds of white-box attacks.Main results. we evaluated the performance of eight adversarial detection approaches on three EEG datasets, three neural networks, and four types of adversarial attacks. Our approach achieved an area under the curve score of up to 99.99% in detecting white-box attacks. Additionally, we assessed the transferability of different adversarial detectors to unknown attacks.Significance. through extensive experiments, we found that white-box attacks may be easily detected, and differences exist in the distributions of different types of adversarial examples. Our work should facilitate understanding the vulnerability of existing BCI models and developing more secure BCIs in the future.

Compact convolutional transformer for subject-independent motor imagery EEG-based BCIs

Motor imagery electroencephalography (EEG) analysis is crucial for the development of effective brain-computer interfaces (BCIs), yet it presents considerable challenges due to the complexity of the data and inter-subject variability. This paper introduces EEGCCT, an application of compact convolutional transformers designed specifically to improve the analysis of motor imagery tasks in EEG. Unlike traditional approaches, EEGCCT model significantly enhances generalization from limited data, effectively addressing a common limitation in EEG datasets. We validate and test our models using the open-source BCI Competition IV datasets 2a and 2b, employing a Leave-One-Subject-Out (LOSO) strategy to ensure subject-independent performance. Our findings demonstrate that EEGCCT not only outperforms conventional models like EEGNet in standard evaluations but also achieves better performance compared to other advanced models such as Conformer, Hybrid s-CViT, and Hybrid t-CViT, while utilizing fewer parameters and achieving an accuracy of 70.12%. Additionally, the paper presents a comprehensive ablation study that includes targeted data augmentation, hyperparameter optimization, and architectural improvements.

DO NOT LOSE SLEEP OVER IT: IMPLANTED BRAIN-COMPUTER INTERFACE FUNCTIONALITY DURING NIGHTTIME IN LATE-STAGE AMYOTROPHIC LATERAL SCLEROSIS

Background and objectives

Brain-computer interfaces (BCIs) hold promise as augmentative and alternative communication technology for people with severe motor and speech impairment (locked-in syndrome) due to neural disease or injury. Although such BCIs should be available 24/7, to enable communication at all times, feasibility of nocturnal BCI use has not been investigated. Here, we addressed this question using data from an individual with amyotrophic lateral sclerosis (ALS) who was implanted with an electrocorticography-based BCI that enabled the generation of click-commands for spelling words and call-caregiver signals.

Methods

We investigated nocturnal dynamics of neural signal features used for BCI control, namely low (LFB: 10-30Hz) and high frequency band power (HFB: 65-95Hz). Additionally, we assessed the nocturnal performance of a BCI decoder that was trained on daytime data by quantifying the number of unintentional BCI activations at night. Finally, we developed and implemented a nightmode decoder that allowed the participant to call a caregiver at night, and assessed its performance.

Results

Power and variance in HFB and LFB were significantly higher at night than during the day in the majority of the nights, with HFB variance being higher in 88% of nights. Daytime decoders caused 245 unintended selection-clicks and 13 unintended caregiver-calls per hour when applied to night data. The developed nightmode decoder functioned error-free in 79% of nights over a period of ±1.5 years, allowing the user to reliably call the caregiver, with unintended activations occurring only once every 12 nights.

Discussion

Reliable nighttime use of a BCI requires decoders that are adjusted to sleep-related signal changes. This demonstration of a reliable BCI nightmode and its long-term use by an individual with advanced ALS underscores the importance of 24/7 BCI reliability.

Trial registration

This trial is registered in clinicaltrials.gov under number NCT02224469 (https://clinicaltrials.gov/study/NCT02224469?term=NCT02224469&rank=1). Date of submission to registry: August 21, 2014. Enrollment of first participant: September 7, 2015.

Filter banks guided correlational convolutional neural network for SSVEPs based BCI classification

Objective.In the field of steady-state visual evoked potential brain computer interfaces (SSVEP-BCIs) research, convolutional neural networks (CNNs) have gradually been proved to be an effective method. Whereas, majority works apply the frequency domain characteristics in long time window to train the network, thus lead to insufficient performance of those networks in short time window. Furthermore, only the frequency domain information for classification lacks of other task-related information.Approach.To address these issues, we propose a time-frequency domain generalized filter-bank convolutional neural network (FBCNN-G) to improve the SSVEP-BCIs classification performance. The network integrates multiple frequency information of electroencephalogram (EEG) with template and predefined prior of sine-cosine signals to perform feature extraction, which contains correlation analyses in both template and signal aspects. Then the classification is performed at the end of the network. In addition, the method proposes the use of filter banks divided into specific frequency bands as pre-filters in the network to fully consider the fundamental and harmonic frequency characteristics of the signal.Main results.The proposed FBCNN-G model is compared with other methods on the public dataset Benchmark. The results manifest that this model has higher accuracy of character recognition accuracy and information transfer rates in several time windows. Particularly, in the 0.2 s time window, the mean accuracy of the proposed method reaches62.02%±5.12%, indicating its superior performance.Significance.The proposed FBCNN-G model is critical for the exploitation of SSVEP-BCIs character recognition models.

Cross-Channel Specific-Mutual Feature Transfer Learning for Motor Imagery EEG Signals Decoding

In recent years, with the rapid development of deep learning, various deep learning frameworks have been widely used in brain-computer interface (BCI) research for decoding motor imagery (MI) electroencephalogram (EEG) signals to understand brain activity accurately. The electrodes, however, record the mixed activities of neurons. If different features are directly embedded in the same feature space, the specific and mutual features of different neuron regions are not considered, which will reduce the expression ability of the feature itself. We propose a cross-channel specific-mutual feature transfer learning (CCSM-FT) network model to solve this problem. The multibranch network extracts the specific and mutual features of brain's multiregion signals. Effective training tricks are used to maximize the distinction between the two kinds of features. Suitable training tricks can also improve the effectiveness of the algorithm compared with novel models. Finally, we transfer two kinds of features to explore the potential of mutual and specific features to enhance the expressive power of the feature and use the auxiliary set to improve identification performance. The experimental results show that the network has a better classification effect in the BCI Competition IV-2a and the HGD datasets.

Case report: Novel use of clinical brain-computer interfaces in recreation programming for an autistic adolescent with co-occurring attention deficit hyperactivity disorder

Background

In recent years, several autistic children and youth have shown interest in Holland Bloorview Kids Rehabilitation Hospital's clinical brain computer interface (BCI) program. Existing literature about BCI use among autistic individuals has focused solely on cognitive skill development and remediation of challenging behaviors. To date, the benefits of recreational BCI programming with autistic children and youth have not been documented.

Purpose

This case report summarizes the experiences of an autistic male adolescent with co-occurring attention deficit hyperactivity disorder using a BCI for recreation and considers possible benefits with this novel user population.

Methods

A single retrospective chart review was completed with parental guardian's consent.

Findings

The participant demonstrated enjoyment in BCI sessions and requested continued opportunities to engage in BCI programming. This enjoyment correlated with improved Canadian Occupational Performance Measure (COPM) scores in BCI programming, outperforming scores from other recreational programs. Additionally, clinicians observed changes in social communication efforts and self-advocacy in this first autistic participant.

Conclusion

The use of brain computer interfaces in recreational programming provides a novel opportunity for engagement for autistic children and youth that may also support skill development.

A multistrategy differential evolution algorithm combined with Latin hypercube sampling applied to a brain-computer interface to improve the effect of node displacement

Injection molding is a common plastic processing technique that allows melted plastic to be injected into a mold through pressure to form differently shaped plastic parts. In injection molding, in-mold electronics (IME) can include various circuit components, such as sensors, amplifiers, and filters. These components can be injected into the mold to form a whole within the melted plastic and can therefore be very easily integrated into the molded part. The brain-computer interface (BCI) is a direct connection pathway between a human or animal brain and an external device. Through BCIs, individuals can use their own brain signals to control these components, enabling more natural and intuitive interactions. In addition, brain-computer interfaces can also be used to assist in medical treatments, such as controlling prosthetic limbs or helping paralyzed patients regain mobility. Brain-computer interfaces can be realized in two ways: invasively and noninvasively, and in this paper, we adopt a noninvasive approach. First, a helmet model is designed according to head shape, and second, a printed circuit film is made to receive EEG signals and an IME injection mold for the helmet plastic parts. In the electronic film, conductive ink is printed to connect each component. However, improper parameterization during the injection molding process can lead to node displacements and residual stress changes in the molded part, which can damage the circuits in the electronic film and affect its performance. Therefore, in this paper, the use of the BCI molding process to ensure that the node displacement reaches the optimal value is studied. Second, the multistrategy differential evolutionary algorithm is used to optimize the injection molding parameters in the process of brain-computer interface formation. The relationship between the injection molding parameters and the actual target value is investigated through Latin hypercubic sampling, and the optimized parameters are compared with the target parameters to obtain the optimal parameter combination. Under the optimal parameters, the node displacement can be optimized from 0.585 to 0.027 mm, and the optimization rate can reach 95.38%. Ultimately, by detecting whether the voltage difference between the output inputs is within the permissible range, the reliability of the brain-computer interface after node displacement optimization can be evaluated.