Search Results (64)

Brain–computer interface (BCI) technology holds promise for improving motor rehabilitation in stroke patients. This review explores the immediate and long-term effects of BCI training, shedding light on the potential benefits and challenges. Clinical studies have demonstrated that BCIs yield significant immediate improvements in motor functions following stroke. Patients can engage in BCI training safely, making it a viable option for rehabilitation. Evidence from single-group studies consistently supports the effectiveness of BCIs in enhancing patients’ performance. Despite these promising findings, the evidence regarding long-term effects remains less robust. Further studies are needed to determine whether BCI-induced changes are permanent or only last for short durations. While evaluating the outcomes of BCI, one must consider that different BCI training protocols may influence functional recovery. The characteristics of some of the paradigms that we discuss are motor imagery-based BCIs, movement-attempt-based BCIs, and brain-rhythm-based BCIs. Finally, we examine studies suggesting that integrating BCIs with other devices, such as those used for functional electrical stimulation, has the potential to enhance recovery outcomes. We conclude that, while BCIs offer immediate benefits for stroke rehabilitation, addressing long-term effects and optimizing clinical implementation remain critical areas for further investigation. Full article

Background: There are only a few available options for early rehabilitation of severely impaired individuals who must remain bedbound, as most exercise paradigms focus on out-of-bed exercises. To enable these individuals to exercise, we developed a novel hybrid rehabilitation system combining a brain–computer interface (BCI), functional electrical stimulation (FES), and a robotic device. Methods: The BCI assessed the presence of a movement-related cortical potential (MRCP) and triggered the administration of FES to produce movement of the lower limb. The exercise trajectory was supported by the robotic device. To demonstrate the system, an experiment was conducted in an out-of-lab setting by ten able-bodied participants. During exercise, the performance of the BCI was assessed, and the participants evaluated the system using the NASA Task Load Index, Intrinsic Motivation Inventory, and by answering a few subjective questions. Results: The BCI reached a true positive rate of 62.6 ± 9.2% and, on average, predicted the movement initiation 595 ± 129 ms prior to the MRCP peak negativity. All questionnaires showed favorable outcomes for the use of the system. Conclusions: The developed system was usable by all participants, but its clinical feasibility is uncertain due to the total time required for setting up the system. Full article

Parkinson’s disease (PD) is a progressive neurological disorder with motor and non-motor symptoms that are inadequately addressed by current pharmacological and surgical therapies. Brain–computer interfaces (BCIs), particularly those based on electroencephalography (eBCIs), provide a promising, non-invasive approach to personalized neurorehabilitation. This narrative review explores the clinical potential of BCIs in PD, discussing signal acquisition, processing, and control paradigms. eBCIs are well-suited for PD due to their portability, safety, and real-time feedback capabilities. Emerging neurophysiological biomarkers—such as beta-band synchrony, phase–amplitude coupling, and altered alpha-band activity—may support adaptive therapies, including adaptive deep brain stimulation (aDBS), as well as motor and cognitive interventions. BCIs may also aid in diagnosis and personalized treatment by detecting these cortical and subcortical patterns associated with motor and cognitive dysfunction in PD. A structured search identified 11 studies involving 64 patients with PD who used BCIs for aDBS, neurofeedback, and cognitive rehabilitation, showing improvements in motor function, cognition, and engagement. Clinical translation requires attention to electrode design and user-centered interfaces. Ethical issues, including data privacy and equitable access, remain critical challenges. As wearable technologies and artificial intelligence evolve, BCIs could shift PD care from intermittent interventions to continuous, brain-responsive therapy, potentially improving patients’ quality of life and autonomy. This review highlights BCIs as a transformative tool in PD management, although more robust clinical evidence is needed. Full article

Motor Imagery (MI) classification plays a crucial role in enhancing the performance of brain–computer interface (BCI) systems, thereby enabling advanced neurorehabilitation and the development of intuitive brain-controlled technologies. However, MI classification using electroencephalography (EEG) is hindered by spatiotemporal variability and the limited interpretability of deep learning (DL) models. To mitigate these challenges, dropout techniques are employed as regularization strategies. Nevertheless, the removal of critical EEG channels, particularly those from the sensorimotor cortex, can result in substantial spatial information loss, especially under limited training data conditions. This issue, compounded by high EEG variability in subjects with poor performance, hinders generalization and reduces the interpretability and clinical trust in MI-based BCI systems. This study proposes a novel framework integrating channel dropout—a variant of Monte Carlo dropout (MCD)—with class activation maps (CAMs) to enhance robustness and interpretability in MI classification. This integration represents a significant step forward by offering, for the first time, a dedicated solution to concurrently mitigate spatiotemporal uncertainty and provide fine-grained neurophysiologically relevant interpretability in motor imagery classification, particularly demonstrating refined spatial attention in challenging low-performing subjects. We evaluate three DL architectures (ShallowConvNet, EEGNet, TCNet Fusion) on a 52-subject MI-EEG dataset, applying channel dropout to simulate structural variability and LayerCAM to visualize spatiotemporal patterns. Results demonstrate that among the three evaluated deep learning models for MI-EEG classification, TCNet Fusion achieved the highest peak accuracy of 74.4% using 32 EEG channels. At the same time, ShallowConvNet recorded the lowest peak at 72.7%, indicating TCNet Fusion’s robustness in moderate-density montages. Incorporating MCD notably improved model consistency and classification accuracy, especially in low-performing subjects where baseline accuracies were below 70%; EEGNet and TCNet Fusion showed accuracy improvements of up to 10% compared to their non-MCD versions. Furthermore, LayerCAM visualizations enhanced with MCD transformed diffuse spatial activation patterns into more focused and interpretable topographies, aligning more closely with known motor-related brain regions and thereby boosting both interpretability and classification reliability across varying subject performance levels. Our approach offers a unified solution for uncertainty-aware, and interpretable MI classification. Full article

Motor Imagery (MI) based Brain Computer Interfaces (BCIs) have promising applications in neurorehabilitation for individuals who have lost mobility and control over parts of their body due to brain injuries, such as stroke patients. Accurately classifying MI tasks is essential for effective BCI performance, but this task remains challenging due to the complex and non-stationary nature of EEG signals. This study aims to improve the classification of left and right-hand MI tasks by utilizing high-resolution time-frequency features extracted from EEG signals, enhanced with deep learning-based data augmentation techniques. We propose a novel deep learning framework named the Generalized Wavelet Transform-based Deep Convolutional Network (GDC-Net), which integrates multiple components. First, EEG signals recorded from the C3, C4, and Cz channels are transformed into detailed time-frequency representations using the Generalized Morse Wavelet Transform (GMWT). The selected features are then expanded using a Deep Convolutional Generative Adversarial Network (DCGAN) to generate additional synthetic data and address data scarcity. Finally, the augmented feature maps data are subsequently fed into a hybrid CNN-LSTM architecture, enabling both spatial and temporal feature learning for improved classification. The proposed approach is evaluated on the BCI Competition IV dataset 2b. Experimental results showed that the mean classification accuracy and Kappa value are 89.24% and 0.784, respectively, making them the highest compared to the state-of-the-art algorithms. The integration of GMWT and DCGAN significantly enhances feature quality and model generalization, thereby improving classification performance. These findings demonstrate that GDC-Net delivers superior MI classification performance by effectively capturing high-resolution time-frequency dynamics and enhancing data diversity. This approach holds strong potential for advancing MI-based BCI applications, especially in assistive and rehabilitation technologies. Full article

Spinal cord injury (SCI) results in a significant loss of motor, sensory, and autonomic function, imposing substantial biosocial and economic burdens. Traditional approaches, such as stem cell therapy and immune modulation, have faced translational challenges, whereas neuromodulation and digital brain–spinal cord interfaces combining brain–computer interface (BCI) technology and epidural spinal cord stimulation (ESCS) to create brain–spine interfaces (BSIs) offer promising alternatives by leveraging residual neural pathways to restore physiological function. This review examines recent advancements in neuromodulation, focusing on the future translation of clinical trial data to clinical practice. We address key considerations, including scalability, patient selection, surgical techniques, postoperative rehabilitation, and ethical implications. By integrating interdisciplinary collaboration, standardized protocols, and patient-centered design, neuromodulation has the potential to revolutionize SCI rehabilitation, reducing long-term disability and enhancing quality of life globally. Full article

Background: Self-controlled motor imagery combined with assistive devices is promising for enhancing neurorehabilitation. This study developed a soft, Flexible Exoskeleton (flexEXO) for finger movements and investigated whether self-controlled motor tasks facilitate stronger cortical activation than externally controlled conditions. Methods: Twenty-one healthy participants performed grasping tasks under four conditions: Self-Controlled Motion (SCC), Other-Controlled Motion (OCC), Self-Controlled Imagery Only (SCIOC), and Other-Controlled Imagery Only (OCIOC). EEG data were recorded, focusing on event-related desynchronization (ERD) in the μ and β bands during imagery and motion and event-related synchronization (ERS) in the β band during feedback. Source localization was performed using eLORETA. Results: Higher μERD and βERD were observed during self-controlled tasks, particularly in the primary motor cortex and supplementary motor area. Externally controlled tasks showed enhanced activation in the inferior parietal lobule and secondary somatosensory cortex. βERS did not differ significantly across conditions. Source localization revealed that self-controlled tasks engaged motor planning and error-monitoring regions more robustly. Conclusions: The flexEXO device and the comparison of brain activity under different conditions provide insights into the neural mechanisms of motor control and have implications for neurorehabilitation. Full article

The ongoing exploration of brain–computer interfaces (BCIs) provides deeper insights into the workings of the human brain. Motor imagery (MI) tasks, such as imagining movements of the tongue, left and right hands, or feet, can be identified through the analysis of electroencephalography (EEG) signals. The development of BCI systems opens up opportunities for their application in assistive devices, neurorehabilitation, and brain stimulation and brain feedback technologies, potentially helping patients to regain the ability to eat and drink without external help, move, or even speak. In this context, the accurate recognition and deciphering of a patient’s imagined intentions is critical for the development of effective BCI systems. Therefore, to distinguish motor tasks in a manner differing from the commonly used methods in this context, we propose a fractal dimension (FD)-based approach, which effectively captures the self-similarity and complexity of EEG signals. For this purpose, all four classes provided in the BCI Competition IV 2a dataset are utilized with nine different combinations of seven FD methods: Katz, Petrosian, Higuchi, box-counting, MFDFA, DFA, and correlation dimension. The resulting features are then used to train five machine learning models: linear, Gaussian, polynomial support vector machine, regression tree, and stochastic gradient descent. As a result, the proposed method obtained top-tier results, achieving 79.2% accuracy when using the Katz vs. box-counting vs. correlation dimension FD combination (KFD vs. BCFD vs. CDFD) classified by LinearSVM, thus outperforming the state-of-the-art TWSB method (achieving 79.1% accuracy). These results demonstrate that fractal dimension features can be applied to achieve higher classification accuracy for online/offline MI-BCIs, when compared to traditional methods. The application of these findings is expected to facilitate the enhancement of motor imagery brain–computer interface systems, which is a key issue faced by neuroscientists. Full article

Background: Recent advancements in machine learning (ML) have significantly influenced the analysis of brain signals, particularly electroencephalography (EEG), enhancing the detection of complex neural patterns. ML enables large-scale data processing, offering novel opportunities for diagnosing and treating mental disorders. However, challenges such as data variability, noise, and model interpretability remain significant. This study reviews the current limitations of EEG-based anxiety detection and explores the potential of advanced AI models, including transformers and VAE-D2GAN, to improve diagnostic accuracy and real-time monitoring. Methods: The paper presents the application of ML algorithms, with a focus on convolutional neural networks (CNN) and recurrent neural networks (RNN), in identifying biomarkers of anxiety disorders and predicting therapy responses. Additionally, it discusses the role of brain–computer interfaces (BCIs) in assisting individuals with disabilities by enabling device control through brain activity. Results: Experimental EEG research on BCI applications was conducted, focusing on motor imagery-based brain activity. Findings indicate that successive training sessions improve signal classification accuracy, emphasizing the need for personalized and adaptive EEG analysis methods. Challenges in BCI usability and technological constraints in EEG processing are also addressed. Conclusions: By integrating ML with EEG analysis, this study highlights the potential for future healthcare applications, including neurorehabilitation, anxiety disorder therapy, and predictive clinical models. Future research should focus on optimizing ML algorithms, enhancing personalization, and addressing ethical concerns related to patient privacy. Full article

Electroencephalography (EEG) remains an essential method for monitoring brain activity, but the limitations of conventional systems due to the complexity of installation and lack of portability have led to the introduction and development of in-ear EEG technology. In-ear EEG is an emerging method of recording electrical activity in the brain and is an innovative concept that offers multiple advantages both from the point of view of the device itself, which is easily portable, and from the user’s point of view, who is more comfortable with it, even in long-term use. One of the fundamental components of this type of device is the electrodes used to capture the EEG signal. This innovative method allows bioelectrical signals to be captured through electrodes integrated into an earpiece, offering significant advantages in terms of comfort, portability, and accessibility. Recent studies have demonstrated that in-ear EEG can record signals qualitatively comparable to scalp EEG, with an optimized signal-to-noise ratio and improved electrode stability. Furthermore, this review provides a comparative synthesis of performance parameters such as signal-to-noise ratio (SNR), common-mode rejection ratio (CMRR), signal amplitude, and comfort, highlighting the strengths and limitations of in-ear EEG systems relative to conventional scalp EEG. This study also introduces a visual model outlining the stages of technological development for in-ear EEG, from initial research to clinical and commercial deployment. Particular attention is given to current innovations in electrode materials and design strategies aimed at balancing biocompatibility, signal fidelity, and anatomical adaptability. This article analyzes the evolution of EEG in the ear, briefly presents the comparative aspects of EEG—EEG in the ear from the perspective of the electrodes used, highlighting the advantages and challenges of using this new technology. It also discusses aspects related to the electrodes used in EEG in the ear: types of electrodes used in EEG in the ear, improvement of contact impedance, and adaptability to the anatomical variability of the ear canal. A comparative analysis of electrode performance in terms of signal quality, long-term stability, and compatibility with use in daily life was also performed. The integration of intra-auricular EEG in wearable devices opens new perspectives for clinical applications, including sleep monitoring, epilepsy diagnosis, and brain–computer interfaces. This study highlights the challenges and prospects in the development of in-ear EEG electrodes, with a focus on integration into wearable devices and the use of biocompatible materials to improve durability and enhance user comfort. Despite its considerable potential, the widespread deployment of in-ear EEG faces challenges such as anatomical variability of the ear canal, optimization of ergonomics, and reduction in motion artifacts. Future research aims to improve device design for long-term monitoring, integrate advanced signal processing algorithms, and explore applications in neurorehabilitation and early diagnosis of neurodegenerative diseases. Full article

(1) Background: Brain–computer interfaces (BCIs) enable direct communication between the brain and external devices using electroencephalography (EEG) signals, offering potential applications in assistive technology and neurorehabilitation. Code-modulated visual evoked potential (cVEP)-based BCIs employ code-pattern-based stimulation to evoke neural responses, which can then be classified to infer user intent. While increasing the number of EEG electrodes across the visual cortex enhances classification accuracy, it simultaneously reduces user comfort and increases setup complexity, duration, and hardware costs. (2) Methods: This online BCI study, involving thirty-eight able-bodied participants, investigated how reducing the electrode count from 16 to 6 affected performance. Three experimental conditions were tested: a baseline 16-electrode configuration, a reduced 6-electrode setup without retraining, and a reduced 6-electrode setup with retraining. (3) Results: Our results indicate that, on average, performance declines with fewer electrodes; nonetheless, retraining restored near-baseline mean Information Transfer Rate (ITR) and accuracy for those participants for whom the system remained functional. The results reveal that for a substantial number of participants, the classification pipeline fails after electrode removal, highlighting individual differences in the cVEP response characteristics or inherent limitations of the classification approach. (4) Conclusions: Ultimately, this suggests that minimal cVEP-BCI electrode setups capable of reliably functioning across all users might only be feasible through other, more flexible classification methods that can account for individual differences. These findings aim to serve as a guideline for what is currently achievable with this common cVEP paradigm and to highlight where future research should focus in order to move closer to a practical and user-friendly system. Full article

Background: The sense of agency (SoA)—the subjective experience of controlling one’s own actions and their consequences—is a fundamental aspect of human cognition, volition, and motor control. Understanding how the SoA arises and is disrupted in neuropsychiatric disorders has significant implications for human–machine interface (HMI) design for neurorehabilitation. Traditional cognitive models of agency often fail to capture its full complexity, especially in dynamic and uncertain environments. Objective: This review synthesizes computational models—particularly predictive coding, Bayesian inference, and optimal control theories—to provide a unified framework for understanding the SoA in both healthy and dysfunctional brains. It aims to demonstrate how these models can inform the design of adaptive HMIs and therapeutic tools by aligning with the brain’s own inference and control mechanisms. Methods: I reviewed the foundational and contemporary literature on predictive coding, Kalman filtering, the Linear–Quadratic–Gaussian (LQG) control framework, and active inference. I explored their integration with neurophysiological mechanisms, focusing on the somato-cognitive action network (SCAN) and its role in sensorimotor integration, intention encoding, and the judgment of agency. Case studies, simulations, and XR-based rehabilitation paradigms using robotic haptics were used to illustrate theoretical concepts. Results: The SoA emerges from hierarchical inference processes that combine top–down motor intentions with bottom–up sensory feedback. Predictive coding frameworks, especially when implemented via Kalman filters and LQG control, provide a mechanistic basis for modeling motor learning, error correction, and adaptive control. Disruptions in these inference processes underlie symptoms in disorders such as functional movement disorder. XR-based interventions using robotic interfaces can restore the SoA by modulating sensory precision and motor predictions through adaptive feedback and suggestion. Computer simulations demonstrate how internal models, and hypnotic suggestions influence state estimation, motor execution, and the recovery of agency. Conclusions: Predictive coding and active inference offer a powerful computational framework for understanding and enhancing the SoA in health and disease. The SCAN system serves as a neural hub for integrating motor plans with cognitive and affective processes. Future work should explore the real-time modulation of agency via biofeedback, simulation, and SCAN-targeted non-invasive brain stimulation. Full article

Background/Objectives: Guided imagery techniques, which include mentally picturing motions or activities to help motor recovery, are an important part of neuroplasticity-based motor therapy in stroke patients. Motor imagery (MI) is a kind of guided imagery in neurorehabilitation that focuses on mentally rehearsing certain motor actions in order to improve performance. This systematic review aims to evaluate the current evidence on guided imagery techniques and identify their therapeutic potential in stroke motor rehabilitation. Methods: Randomized controlled trials (RCTs) published in the English language were identified from an online search of PubMed, Web of Science, Embase, EBSCOhost, and Scopus databases without a specific search time frame. The inclusion criteria take into account guided imagery interventions and evaluate their impact on motor recovery through validated clinical, neurophysiological, or functional assessments. This review has been registered on Open OSF with the following number: DOI 10.17605/OSF.IO/3D7MF. Results: This review synthesized 41 RCTs on MI in stroke rehabilitation, with 996 participants in the intervention group and 757 in the control group (average age 50–70, 35% female). MI showed advantages for gait, balance, and upper limb function; however, the RoB 2 evaluation revealed ‘some concerns’ related to allocation concealment, blinding, and selective reporting issues. Integrating MI with gait training or action observation (AO) seems to improve motor recovery, especially in balance and walking. Technological methods like brain–computer interfaces (BCIs) and hybrid models that combine MI with circuit training hold potential for enhancing functional mobility and motor results. Conclusions: Guided imagery shows promise as a beneficial adjunct in stroke rehabilitation, with the potential to improve motor recovery across several domains such as gait, upper limb function, and balance. Full article

Background/Objectives: Motor neurorehabilitation can be realized by gradually learning diverse motor imagery (MI) tasks. EEG-based brain-computer interfaces (BCIs) provide an effective solution. Nevertheless, existing MI decoding methods cannot balance plasticity for unseen tasks and stability for old tasks. This paper proposes a generative diffusion-based task Incremental Learning (IL) method called GD-TIL. Methods: First, data augmentation is employed to increase data diversity by segmenting and recombining EEG signals. Second, to capture temporal-spatial features (TSFs) from different temporal resolutions, a multi-scale temporal-spatial feature extractor (MTSFE) is developed via integrating multiscale temporal-spatial convolutions, a dual-branch pooling operation, multiple multi-head self-attention mechanisms, and a dynamic convolutional encoder. The proposed self-supervised task generalization (SSTG) mechanism introduces a regularization constraint to guide MTSFE and unified classifier updating, which combines labels and semantic similarity between the augmentation with original views to enhance model generalizability for unseen tasks. In the IL phase, a prototype-guided generative replay module (PGGR) is used to generate old tasks’ TSFs by training a lightweight diffusion model based on the prototype and label of each task. Furthermore, the generated TSF is merged with a new TSF to fine-tune the convolutional encoder and update the classifier and PGGR. Finally, GD-TIL is evaluated on a self-collected ADL-MI dataset with two MI pairs and a public dataset with four MI tasks. Results: The continuous decoding accuracy reaches 80.20% and 81.32%, respectively. The experimental results exhibit the excellent plasticity and stability of GD-TIL, even beating the state-of-the-art IL methods. Conclusions: Our work illustrates the potential of MI-based BCI and generative AI for continuous neurorehabilitation. Full article

Enhancing motor disability assessment and its imagery classification is a significant concern in contemporary medical practice, necessitating reliable solutions to improve patient outcomes. One promising avenue is the use of brain–computer interfaces (BCIs), which establish a direct communication pathway between users and machines. This technology holds the potential to revolutionize human–machine interaction, especially for individuals diagnosed with motor disabilities. Despite this promise, extracting reliable control signals from noisy brain data remains a critical challenge. In this paper, we introduce a novel approach leveraging the collaborative synergy of five convolutional neural network (CNN) models to improve the classification accuracy of motor imagery tasks, which are essential components of BCI systems. Our method demonstrates exceptional performance, achieving an accuracy of 79.44% on the BCI Competition IV 2a dataset, surpassing existing state-of-the-art techniques in using multiple CNN models. This advancement offers significant promise for enhancing the efficacy and versatility of BCIs in a wide range of real-world applications, from assistive technologies to neurorehabilitation, thereby providing robust solutions for individuals with motor disabilities. Full article