Intra- and Inter-subject Variability in EEG-Based Sensorimotor Brain Computer Interface: A Review

The Constrained Disorder Principle Overcomes the Challenges of Methods for Assessing Uncertainty in Biological Systems

Different disciplines are developing various methods for determining and dealing with uncertainties in complex systems. The constrained disorder principle (CDP) accounts for the randomness, variability, and uncertainty that characterize biological systems and are essential for their proper function. Per the CDP, intrinsic unpredictability is mandatory for the dynamicity of biological systems under continuously changing internal and external perturbations. The present paper describes some of the parameters and challenges associated with uncertainty and randomness in biological systems and presents methods for quantifying them. Modeling biological systems necessitates accounting for the randomness, variability, and underlying uncertainty of systems in health and disease. The CDP provides a scheme for dealing with uncertainty in biological systems and sets the basis for using them. This paper presents the CDP-based second-generation artificial intelligence system that incorporates variability to improve the effectiveness of medical interventions. It describes the use of the digital pill that comprises algorithm-based personalized treatment regimens regulated by closed-loop systems based on personalized signatures of variability. The CDP provides a method for using uncertainties in complex systems in an outcome-based manner.

Improving subject transfer in EEG classification with divergence estimation

Objective. Classification models for electroencephalogram (EEG) data show a large decrease in performance when evaluated on unseen test subjects. We improve performance using new regularization techniques during model training.Approach. We propose several graphical models to describe an EEG classification task. From each model, we identify statistical relationships that should hold true in an idealized training scenario (with infinite data and a globally-optimal model) but that may not hold in practice. We design regularization penalties to enforce these relationships in two stages. First, we identify suitable proxy quantities (divergences such as Mutual Information and Wasserstein-1) that can be used to measure statistical independence and dependence relationships. Second, we provide algorithms to efficiently estimate these quantities during training using secondary neural network models.Main results. We conduct extensive computational experiments using a large benchmark EEG dataset, comparing our proposed techniques with a baseline method that uses an adversarial classifier. We first show the performance of each method across a wide range of hyperparameters, demonstrating that each method can be easily tuned to yield significant benefits over an unregularized model. We show that, using ideal hyperparameters for all methods, our first technique gives significantly better performance than the baseline regularization technique. We also show that, across hyperparameters, our second technique gives significantly more stable performance than the baseline. The proposed methods require only a small computational cost at training time that is equivalent to the cost of the baseline.Significance. The high variability in signal distribution between subjects means that typical approaches to EEG signal modeling often require time-intensive calibration for each user, and even re-calibration before every use. By improving the performance of population models in the most stringent case of zero-shot subject transfer, we may help reduce or eliminate the need for model calibration.

Free Will as Defined by the Constrained Disorder Principle: a Restricted, Mandatory, Personalized, Regulated Process for Decision-Making

The concept of free will has challenged physicists, biologists, philosophers, and other professionals for decades. The constrained disorder principle (CDP) is a fundamental law that defines systems according to their inherent variability. It provides mechanisms for adapting to dynamic environments. This work examines the CDP's perspective of free will concerning various free will theories. Per the CDP, systems lack intentions, and the "freedom" to select and act is built into their design. The "freedom" is embedded within the response range determined by the boundaries of the systems' variability. This built-in and self-generating mechanism enables systems to cope with perturbations. According to the CDP, neither dualism nor an unknown metaphysical apparatus dictates choices. Brain variability facilitates cognitive adaptation to complex, unpredictable situations across various environments. Human behaviors and decisions reflect an underlying physical variability in the brain and other organs for dealing with unpredictable noises. Choices are not predetermined but reflect the ongoing adaptation processes to dynamic prssu½res. Malfunctions and disease states are characterized by inappropriate variability, reflecting an inability to respond adequately to perturbations. Incorporating CDP-based interventions can overcome malfunctions and disease states and improve decision processes. CDP-based second-generation artificial intelligence platforms improve interventions and are being evaluated to augment personal development, wellness, and health.

Clustering for mitigating subject variability in driving fatigue classification using electroencephalography source-space functional connectivity features

Objective.While Electroencephalography (EEG)-based driver fatigue state classification models have demonstrated effectiveness, their real-world application remains uncertain. The substantial variability in EEG signals among individuals poses a challenge in developing a universal model, often necessitating retraining with the introduction of new subjects. However, obtaining sufficient data for retraining, especially fatigue data for new subjects, is impractical in real-world settings.Approach.In response to these challenges, this paper introduces a hybrid solution for fatigue detection that combines clustering with classification. Unsupervised clustering groups subjects based on their EEG functional connectivity (FC) in an alert state, and classification models are subsequently applied to each cluster for predicting alert and fatigue states.Main results. Results indicate that classification on clusters achieves higher accuracy than scenarios without clustering, suggesting successful grouping of subjects with similar FC characteristics through clustering, thereby enhancing the classification process.Significance.Furthermore, the proposed hybrid method ensures a practical and realistic retraining process, improving the adaptability and effectiveness of the fatigue detection system in real-world applications.

The Co-Piloting Model for Using Artificial Intelligence Systems in Medicine: Implementing the Constrained-Disorder-Principle-Based Second-Generation System

The development of artificial intelligence (AI) and machine learning (ML)-based systems in medicine is growing, and these systems are being used for disease diagnosis, drug development, and treatment personalization. Some of these systems are designed to perform activities that demand human cognitive function. However, use of these systems in routine care by patients and caregivers lags behind expectations. This paper reviews several challenges that healthcare systems face and the obstacles of integrating digital systems into routine care. This paper focuses on integrating digital systems with human physicians. It describes second-generation AI systems designed to move closer to biology and reduce complexity, augmenting but not replacing physicians to improve patient outcomes. The constrained disorder principle (CDP) defines complex biological systems by their degree of regulated variability. This paper describes the CDP-based second-generation AI platform, which is the basis for the Digital Pill that is humanizing AI by moving closer to human biology via using the inherent variability of biological systems for improving outcomes. This system augments physicians, assisting them in decision-making to improve patients' responses and adherence but not replacing healthcare providers. It restores the efficacy of chronic drugs and improves adherence while generating data-driven therapeutic regimens. While AI can substitute for many medical activities, it is unlikely to replace human physicians. Human doctors will continue serving patients with capabilities augmented by AI. The described co-piloting model better reflects biological pathways and provides assistance to physicians for better care.

Utility of Electroencephalograms for Enhancing Clinical Care and Rehabilitation of Children with Acquired Brain Injury

The objective of this literature review was to present evidence from recent studies and applications focused on employing electroencephalogram (EEG) monitoring and methodological approaches during the rehabilitation of children with acquired brain injuries and their related effects. We describe acquired brain injury (ABI) as one of the most common reasons for cognitive and motor disabilities in children that significantly impact their safety, independence, and overall quality of life. These disabilities manifest as dysfunctions in cognition, gait, balance, upper-limb coordination, and hand dexterity. Rehabilitation treatment aims to restore and optimize these impaired functions to help children regain autonomy and enhance their quality of life. Recent advancements in monitoring technologies such as EEG measurements are increasingly playing a role in clinical diagnosis and management. A significant advantage of incorporating EEG technology in pediatric rehabilitation is its ability to provide continuous and objective quantitative monitoring of a child's neurological status. This allows for the real-time assessment of improvement or deterioration in brain function, including, but not limited to, a significant impact on motor function. EEG monitoring enables healthcare providers to tailor and adjust interventions-both pharmacological and rehabilitative-based on the child's current neurological status.

SpeechBrain-MOABB: An open-source Python library for benchmarking deep neural networks applied to EEG signals

Deep learning has revolutionized EEG decoding, showcasing its ability to outperform traditional machine learning models. However, unlike other fields, EEG decoding lacks comprehensive open-source libraries dedicated to neural networks. Existing tools (MOABB and braindecode) prevent the creation of robust and complete decoding pipelines, as they lack support for hyperparameter search across the entire pipeline, and are sensitive to fluctuations in results due to network random initialization. Furthermore, the absence of a standardized experimental protocol exacerbates the reproducibility crisis in the field. To address these limitations, we introduce SpeechBrain-MOABB, a novel open-source toolkit carefully designed to facilitate the development of a comprehensive EEG decoding pipeline based on deep learning. SpeechBrain-MOABB incorporates a complete experimental protocol that standardizes critical phases, such as hyperparameter search and model evaluation. It natively supports multi-step hyperparameter search for finding the optimal hyperparameters in a high-dimensional space defined by the entire pipeline, and multi-seed training and evaluation for obtaining performance estimates robust to the variability caused by random initialization. SpeechBrain-MOABB outperforms other libraries, including MOABB and braindecode, with accuracy improvements of 14.9% and 25.2% (on average), respectively. By enabling easy-to-use and easy-to-share decoding pipelines, our toolkit can be exploited by neuroscientists for decoding EEG with neural networks in a replicable and trustworthy way.

Using the Constrained Disorder Principle to Navigate Uncertainties in Biology and Medicine: Refining Fuzzy Algorithms

Uncertainty in biology refers to situations in which information is imperfect or unknown. Variability, on the other hand, is measured by the frequency distribution of observed data. Biological variability adds to the uncertainty. The Constrained Disorder Principle (CDP) defines all systems in the universe by their inherent variability. According to the CDP, systems exhibit a degree of variability necessary for their proper function, allowing them to adapt to changes in their environments. Per the CDP, while variability differs from uncertainty, it can be viewed as a regulated mechanism for efficient functionality rather than uncertainty. This paper explores the various aspects of un-certainties in biology. It focuses on using CDP-based platforms for refining fuzzy algorithms to address some of the challenges associated with biological and medical uncertainties. Developing a fuzzy decision tree that considers the natural variability of systems can help minimize uncertainty. This method can reveal previously unidentified classes, reduce the number of unknowns, improve the accuracy of modeling results, and generate algorithm outputs that are more biologically and clinically relevant.

Harnessing Variability Signatures and Biological Noise May Enhance Immunotherapies' Efficacy and Act as Novel Biomarkers for Diagnosing and Monitoring Immune-Associated Disorders

Lack of response to immunotherapies poses a significant challenge in treating immune-mediated disorders and cancers. While the mechanisms associated with poor responsiveness are not well defined and change between and among subjects, the current methods for overcoming the loss of response are insufficient. The Constrained Disorder Principle (CDP) explains biological systems based on their inherent variability, bounded by dynamic boundaries that change in response to internal and external perturbations. Inter and intra-subject variability characterize the immune system, making it difficult to provide a single therapeutic regimen to all patients and even the same patients over time. The dynamicity of the immune variability is also a significant challenge for personalizing immunotherapies. The CDP-based second-generation artificial intelligence system is an outcome-based dynamic platform that incorporates personalized variability signatures into the therapeutic regimen and may provide methods for improving the response and overcoming the loss of response to treatments. The signatures of immune variability may also offer a method for identifying new biomarkers for early diagnosis, monitoring immune-related disorders, and evaluating the response to treatments.

The Role of Interoceptive Sensitivity and Hypnotizability in Motor Imagery

Motor imagery (MI) requires the mental representation of the body, obtained by integrating exteroceptive and interoceptive information. This study aimed to investigate the role of interoceptive sensitivity (IS) in MI performed through visual and kinesthetic modalities by participants with low (lows, N = 26; SHSS: A, M + SD: 1.00 + 1.52), medium (mediums, N = 11; SHSS: A, 6.00 + 0.77) and high hypnotizability scores (highs, N = 16; SHSS:A, 9.75 + 1.24), as measured by the Stanford Hypnotic Susceptibility Scale: Form A. The three groups displayed different MI abilities and IS levels. The efficacy of MI was measured using the chronometric index and self-reported experience, while IS was measured using the Multidimensional Assessment of Interoceptive Awareness (MAIA) questionnaire. Alpha and beta power spectrum densities (PSDs) were extracted from the EEG signals acquired during baseline, actual movement and visually and kinesthetically imagined movements. The chronometric indices do not reveal significant differences between groups and imagery modalities. The self-report MI efficacy indicates better kinesthetic imagery in highs and mediums than in lows, and no modality difference among lows. The MAIA dimensions sustain the differences in subjective experience and almost all the EEG differences. The latter are slightly different in highs, mediums and lows. This is the first report of the major role played by IS in MI and strongly supports the theory of embodied cognition.

Evaluating the real-world usability of BCI control systems with augmented reality: a user study protocol

Brain-computer interfaces (BCI) enable users to control devices through their brain activity. Motor imagery (MI), the neural activity resulting from an individual imagining performing a movement, is a common control paradigm. This study introduces a user-centric evaluation protocol for assessing the performance and user experience of an MI-based BCI control system utilizing augmented reality. Augmented reality is employed to enhance user interaction by displaying environment-aware actions, and guiding users on the necessary imagined movements for specific device commands. One of the major gaps in existing research is the lack of comprehensive evaluation methodologies, particularly in real-world conditions. To address this gap, our protocol combines quantitative and qualitative assessments across three phases. In the initial phase, the BCI prototype's technical robustness is validated. Subsequently, the second phase involves a performance assessment of the control system. The third phase introduces a comparative analysis between the prototype and an alternative approach, incorporating detailed user experience evaluations through questionnaires and comparisons with non-BCI control methods. Participants engage in various tasks, such as object sorting, picking and placing, and playing a board game using the BCI control system. The evaluation procedure is designed for versatility, intending applicability beyond the specific use case presented. Its adaptability enables easy customization to meet the specific user requirements of the investigated BCI control application. This user-centric evaluation protocol offers a comprehensive framework for iterative improvements to the BCI prototype, ensuring technical validation, performance assessment, and user experience evaluation in a systematic and user-focused manner.

Exploring the frontier: Transformer-based models in EEG signal analysis for brain-computer interfaces

This review systematically explores the application of transformer-based models in EEG signal processing and brain-computer interface (BCI) development, with a distinct focus on ensuring methodological rigour and adhering to empirical validations within the existing literature. By examining various transformer architectures, such as the Temporal Spatial Transformer Network (TSTN) and EEG Conformer, this review delineates their capabilities in mitigating challenges intrinsic to EEG data, such as noise and artifacts, and their subsequent implications on decoding and classification accuracies across disparate mental tasks. The analytical scope extends to a meticulous examination of attention mechanisms within transformer models, delineating their role in illuminating critical temporal and spatial EEG features and facilitating interpretability in model decision-making processes. The discourse additionally encapsulates emerging works that substantiate the efficacy of transformer models in noise reduction of EEG signals and diversifying applications beyond the conventional motor imagery paradigm. Furthermore, this review elucidates evident gaps and propounds exploratory avenues in the applications of pre-trained transformers in EEG analysis and the potential expansion into real-time and multi-task BCI applications. Collectively, this review distils extant knowledge, navigates through the empirical findings, and puts forward a structured synthesis, thereby serving as a conduit for informed future research endeavours in transformer-enhanced, EEG-based BCI systems.

Deep learning-based auditory attention decoding in listeners with hearing impairment. J Neural Eng 2024;21:036022. [PMID: 38729132 DOI: 10.1088/1741-2552/ad49d7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

Objective.This study develops a deep learning (DL) method for fast auditory attention decoding (AAD) using electroencephalography (EEG) from listeners with hearing impairment (HI). It addresses three classification tasks: differentiating noise from speech-in-noise, classifying the direction of attended speech (left vs. right) and identifying the activation status of hearing aid noise reduction algorithms (OFF vs. ON). These tasks contribute to our understanding of how hearing technology influences auditory processing in the hearing-impaired population.Approach.Deep convolutional neural network (DCNN) models were designed for each task. Two training strategies were employed to clarify the impact of data splitting on AAD tasks: inter-trial, where the testing set used classification windows from trials that the training set had not seen, and intra-trial, where the testing set used unseen classification windows from trials where other segments were seen during training. The models were evaluated on EEG data from 31 participants with HI, listening to competing talkers amidst background noise.Main results.Using 1 s classification windows, DCNN models achieve accuracy (ACC) of 69.8%, 73.3% and 82.9% and area-under-curve (AUC) of 77.2%, 80.6% and 92.1% for the three tasks respectively on inter-trial strategy. In the intra-trial strategy, they achieved ACC of 87.9%, 80.1% and 97.5%, along with AUC of 94.6%, 89.1%, and 99.8%. Our DCNN models show good performance on short 1 s EEG samples, making them suitable for real-world applications. Conclusion: Our DCNN models successfully addressed three tasks with short 1 s EEG windows from participants with HI, showcasing their potential. While the inter-trial strategy demonstrated promise for assessing AAD, the intra-trial approach yielded inflated results, underscoring the important role of proper data splitting in EEG-based AAD tasks.Significance.Our findings showcase the promising potential of EEG-based tools for assessing auditory attention in clinical contexts and advancing hearing technology, while also promoting further exploration of alternative DL architectures and their potential constraints.

A causal perspective on brainwave modeling for brain-computer interfaces

Objective. Machine learning (ML) models have opened up enormous opportunities in the field of brain-computer Interfaces (BCIs). Despite their great success, they usually face severe limitations when they are employed in real-life applications outside a controlled laboratory setting.Approach. Mixing causal reasoning, identifying causal relationships between variables of interest, with brainwave modeling can change one's viewpoint on some of these major challenges which can be found in various stages in the ML pipeline, ranging from data collection and data pre-processing to training methods and techniques.Main results. In this work, we employ causal reasoning and present a framework aiming to breakdown and analyze important challenges of brainwave modeling for BCIs.Significance. Furthermore, we present how general ML practices as well as brainwave-specific techniques can be utilized and solve some of these identified challenges. And finally, we discuss appropriate evaluation schemes in order to measure these techniques' performance and efficiently compare them with other methods that will be developed in the future.

On the role of generative artificial intelligence in the development of brain-computer interfaces

Since their inception more than 50 years ago, Brain-Computer Interfaces (BCIs) have held promise to compensate for functions lost by people with disabilities through allowing direct communication between the brain and external devices. While research throughout the past decades has demonstrated the feasibility of BCI to act as a successful assistive technology, the widespread use of BCI outside the lab is still beyond reach. This can be attributed to a number of challenges that need to be addressed for BCI to be of practical use including limited data availability, limited temporal and spatial resolutions of brain signals recorded non-invasively and inter-subject variability. In addition, for a very long time, BCI development has been mainly confined to specific simple brain patterns, while developing other BCI applications relying on complex brain patterns has been proven infeasible. Generative Artificial Intelligence (GAI) has recently emerged as an artificial intelligence domain in which trained models can be used to generate new data with properties resembling that of available data. Given the enhancements observed in other domains that possess similar challenges to BCI development, GAI has been recently employed in a multitude of BCI development applications to generate synthetic brain activity; thereby, augmenting the recorded brain activity. Here, a brief review of the recent adoption of GAI techniques to overcome the aforementioned BCI challenges is provided demonstrating the enhancements achieved using GAI techniques in augmenting limited EEG data, enhancing the spatiotemporal resolution of recorded EEG data, enhancing cross-subject performance of BCI systems and implementing end-to-end BCI applications. GAI could represent the means by which BCI would be transformed into a prevalent assistive technology, thereby improving the quality of life of people with disabilities, and helping in adopting BCI as an emerging human-computer interaction technology for general use.

Subject-independent meta-learning framework towards optimal training of EEG-based classifiers

Advances in deep learning have shown great promise towards the application of performing high-accuracy Electroencephalography (EEG) signal classification in a variety of tasks. However, many EEG-based datasets are often plagued by the issue of high inter-subject signal variability. Robust deep learning models are notoriously difficult to train under such scenarios, often leading to subpar or widely varying performance across subjects under the leave-one-subject-out paradigm. Recently, the model agnostic meta-learning framework was introduced as a way to increase the model's ability to generalize towards new tasks. While the original framework focused on task-based meta-learning, this research aims to show that the meta-learning methodology can be modified towards subject-based signal classification while maintaining the same task objectives and achieve state-of-the-art performance. Namely, we propose the novel implementation of a few/zero-shot subject-independent meta-learning framework towards multi-class inner speech and binary class motor imagery classification. Compared to current subject-adaptive methods which utilize large number of labels from the target, the proposed framework shows its effectiveness in training zero-calibration and few-shot models for subject-independent EEG classification. The proposed few/zero-shot subject-independent meta-learning mechanism performs well on both small and large datasets and achieves robust, generalized performance across subjects. The results obtained shows a significant improvement over the current state-of-the-art, with the binary class motor imagery achieving 88.70% and the accuracy of multi-class inner speech achieving an average of 31.15%. Codes will be made available to public upon publication.

Automated mood disorder symptoms monitoring from multivariate time-series sensory data: getting the full picture beyond a single number

Mood disorders (MDs) are among the leading causes of disease burden worldwide. Limited specialized care availability remains a major bottleneck thus hindering pre-emptive interventions. MDs manifest with changes in mood, sleep, and motor activity, observable in ecological physiological recordings thanks to recent advances in wearable technology. Therefore, near-continuous and passive collection of physiological data from wearables in daily life, analyzable with machine learning (ML), could mitigate this problem, bringing MDs monitoring outside the clinician's office. Previous works predict a single label, either the disease state or a psychometric scale total score. However, clinical practice suggests that the same label may underlie different symptom profiles, requiring specific treatments. Here we bridge this gap by proposing a new task: inferring all items in HDRS and YMRS, the two most widely used standardized scales for assessing MDs symptoms, using physiological data from wearables. To that end, we develop a deep learning pipeline to score the symptoms of a large cohort of MD patients and show that agreement between predictions and assessments by an expert clinician is clinically significant (quadratic Cohen's κ and macro-average F1 score both of 0.609). While doing so, we investigate several solutions to the ML challenges associated with this task, including multi-task learning, class imbalance, ordinal target variables, and subject-invariant representations. Lastly, we illustrate the importance of testing on out-of-distribution samples.

Modeling Complex EEG Data Distribution on the Riemannian Manifold Toward Outlier Detection and Multimodal Classification

OBJECTIVE

The usage of Riemannian geometry for Brain-computer interfaces (BCIs) has gained momentum in recent years. Most of the machine learning techniques proposed for Riemannian BCIs consider the data distribution on a manifold to be unimodal. However, the distribution is likely to be multimodal rather than unimodal since high-data variability is a crucial limitation of electroencephalography (EEG). In this paper, we propose a novel data modeling method for considering complex data distributions on a Riemannian manifold of EEG covariance matrices, aiming to improve BCI reliability.

METHODS

Our method, Riemannian spectral clustering (RiSC), represents EEG covariance matrix distribution on a manifold using a graph with proposed similarity measurement based on geodesic distances, then clusters the graph nodes through spectral clustering. This allows flexibility to model both a unimodal and a multimodal distribution on a manifold. RiSC can be used as a basis to design an outlier detector named outlier detection Riemannian spectral clustering (odenRiSC) and a multimodal classifier named multimodal classifier Riemannian spectral clustering (mcRiSC). All required parameters of odenRiSC/mcRiSC are selected in data-driven manner. Moreover, there is no need to pre-set a threshold for outlier detection and the number of modes for multimodal classification.

RESULTS

The experimental evaluation revealed odenRiSC can detect EEG outliers more accurately than existing methods and mcRiSC outperformed the standard unimodal classifier, especially on high-variability datasets.

CONCLUSION

odenRiSC/mcRiSC are anticipated to contribute to making real-life BCIs outside labs and neuroergonomics applications more robust.

SIGNIFICANCE

RiSC can work as a robust EEG outlier detector and multimodal classifier.

Multimodal pre-screening can predict BCI performance variability: A novel subject-specific experimental scheme

BACKGROUND

Brain-computer interface (BCI) systems currently lack the required robustness for long-term daily use due to inter- and intra-subject performance variability. In this study, we propose a novel personalized scheme for a multimodal BCI system, primarily using functional near-infrared spectroscopy (fNIRS) and electroencephalography (EEG), to identify, predict, and compensate for factors affecting competence-related and interfering factors associated with performance.

METHOD

11 (out of 13 recruited) participants, including five participants with motor deficits, completed four sessions on average. During the training sessions, the subjects performed a short pre-screening phase, followed by three variations of a novel visou-mental (VM) protocol. Features extracted from the pre-screening phase were used to construct predictive platforms using stepwise multivariate linear regression (MLR) models. In the test sessions, we employed a task-correction phase where our predictive models were used to predict the ideal task variation to maximize performance, followed by an interference-correction phase. We then investigated the associations between predicted and actual performances and evaluated the outcome of correction strategies.

RESULT

The predictive models resulted in respective adjusted R-squared values of 0.942, 0.724, and 0.939 for the first, second, and third variation of the task, respectively. The statistical analyses showed significant associations between the performances predicted by predictive models and the actual performances for the first two task variations, with rhos of 0.7289 (p-value = 0.011) and 0.6970 (p-value = 0.017), respectively. For 81.82 % of the subjects, the task/workload correction stage correctly determined which task variation provided the highest accuracy, with an average performance gain of 5.18 % when applying the correction strategies.

CONCLUSION

Our proposed method can lead to an integrated multimodal predictive framework to compensate for BCI performance variability, particularly, for people with severe motor deficits.

Variability in exercise is linked to improved age-related dysfunctions: A potential role for the constrained-disorder principle-based second-generation artificial intelligence system

Objective: Regular physical activity (PA) promotes mental and physical health. Nevertheless, inactivity is a worldwide pandemic, and methods to augment exercise benefits are required. The constrained disorder principle (CDP) characterizes biological systems based on their inherent variability. We aimed to investigate the association between intra-individual variability in PA and disability among non-athlete adults. Methods: In this retrospective analysis of the longitudinal SHARE survey, we included non-disabled adults aged >50 with at least six visits over 14 years. Self-reported PA frequency was documented bi- to triennially. Low PA intensity was defined as vigorous PA frequency less than once a week. Stable PA was described as an unchanged PA intensity in all consecutive middle observations. The primary outcome was defined as a physical limitation in everyday activities at the end of the survey. Secondary outcomes were cognitive functions, including short-term memory, long-term memory, and verbal fluency. Results: The study included 2,049 non-disabled adults with a mean age of 53 and 49.1% women. In the initially high PA intensity group, variability in PA was associated with increased physical disability prevalence (23.3% vs. 33.2%, stable vs. unstable PA ; P<0.01; adjusted P<0.01). In the initially low PA intensity group, variability was associated with a reduced physical disability (45.6% vs. 33.3%, stable vs. unstable PA ; P=0.02; adjusted P=0.03). There were no statistically significant differences in cognitive parameters between the groups. Among individuals with the same low PA intensity at the beginning and end of follow-up, variability was associated with reduced physical disability (56.9% vs. 36.5%, stable vs. unstable PA ; P=0.02; adjusted P=0.04) and improved short-term memory (score change: -0.28 vs. +0.29, stable vs. unstable PA ; P=0.05). Conclusion: Incorporating variability into PA regimens of inactive adults may enhance their physical and cognitive benefits.

Deep Unsupervised Representation Learning for Feature-Informed EEG Domain Extraction

In electroencephalography (EEG) classification paradigms, data from a target subject is often difficult to obtain, leading to difficulties in training a robust deep learning network. Transfer learning and their variations are effective tools in improving such models suffering from lack of data. However, many of the proposed variations and deep models often rely on a single assumed distribution to represent the latent features which may not scale well due to inter- and intra-subject variations in signals. This leads to significant instability in individual subject decoding performances. The presence of non-trivial domain differences between different sets of training or transfer learning data causes poorer model generalization towards the target subject. However, the detection of these domain differences is often difficult to perform due to the ill-defined nature of the EEG domain features. This study proposes a novel inference model, the Joint Embedding Variational Autoencoder, that offers conditionally tighter approximation of the estimated spatiotemporal feature distribution through the use of jointly optimised variational autoencoders to achieve optimizable data dependent inputs as an additional variable for improved overall model optimisation and scaling without sacrificing model tightness. To learn the variational bound, we show that maximising the marginal log-likelihood of only the second embedding section is required to achieve conditionally tighter lower bounds. Furthermore, we show that this model provides state-of-the-art EEG data reconstruction and deep feature extraction. The extracted domains of the EEG signals across each subject displays the rationale as to why there exists disparity between subjects' adaptation efficacy.

Culture shapes spontaneous brain dynamics - Shared versus idiosyncratic neural features among Chinese versus Canadian subjects

Environmental factors, such as culture, are known to shape individual variation in brain activity including spontaneous activity, but less is known about their population-level effects. Eastern and Western cultures differ strongly in their cultural norms about relationships between individuals. For example, the collectivism, interdependence and tightness of Eastern cultures relative to the individualism, independence and looseness of Western cultures, promote interpersonal connectedness and coordination. Do such cultural contexts therefore influence the group-level variability of their cultural members' spontaneous brain activity? Using novel methods adapted from studies of inter-subject neural synchrony, we compare the group-level variability of resting state EEG dynamics in Chinese and Canadian samples. We observe that Chinese subjects show significantly higher inter-subject correlation and lower inter-subject distance in their EEG power spectra than Canadian subjects, as well as lower variability in theta power and alpha peak frequency. We demonstrate, for the first time, different relationships among subjects' resting state brain dynamics in Chinese and Canadian samples. These results point to more idiosyncratic neural dynamics among Canadian participants, compared with more shared neural features in Chinese participants.

Group-level brain decoding with deep learning

Decoding brain imaging data are gaining popularity, with applications in brain-computer interfaces and the study of neural representations. Decoding is typically subject-specific and does not generalise well over subjects, due to high amounts of between subject variability. Techniques that overcome this will not only provide richer neuroscientific insights but also make it possible for group-level models to outperform subject-specific models. Here, we propose a method that uses subject embedding, analogous to word embedding in natural language processing, to learn and exploit the structure in between-subject variability as part of a decoding model, our adaptation of the WaveNet architecture for classification. We apply this to magnetoencephalography data, where 15 subjects viewed 118 different images, with 30 examples per image; to classify images using the entire 1 s window following image presentation. We show that the combination of deep learning and subject embedding is crucial to closing the performance gap between subject- and group-level decoding models. Importantly, group models outperform subject models on low-accuracy subjects (although slightly impair high-accuracy subjects) and can be helpful for initialising subject models. While we have not generally found group-level models to perform better than subject-level models, the performance of group modelling is expected to be even higher with bigger datasets. In order to provide physiological interpretation at the group level, we make use of permutation feature importance. This provides insights into the spatiotemporal and spectral information encoded in the models. All code is available on GitHub (https://github.com/ricsinaruto/MEG-group-decode).

BrainWave-Scattering Net: a lightweight network for EEG-based motor imagery recognition

Objective.Brain-computer interfaces (BCIs) enable a direct communication of the brain with the external world, using one's neural activity, measured by electroencephalography (EEG) signals. In recent years, convolutional neural networks (CNNs) have been widely used to perform automatic feature extraction and classification in various EEG-based tasks. However, their undeniable benefits are counterbalanced by the lack of interpretability properties as well as the inability to perform sufficiently when only limited amount of training data is available.Approach.In this work, we introduce a novel, lightweight, fully-learnable neural network architecture that relies on Gabor filters to delocalize EEG signal information into scattering decomposition paths along frequency and slow-varying temporal modulations.Main results.We utilize our network in two distinct modeling settings, for building either a generic (training across subjects) or a personalized (training within a subject) classifier.Significance.In both cases, using two different publicly available datasets and one in-house collected dataset, we demonstrate high performance for our model with considerably less number of trainable parameters as well as shorter training time compared to other state-of-the-art deep architectures. Moreover, our network demonstrates enhanced interpretability properties emerging at the level of the temporal filtering operation and enables us to train efficient personalized BCI models with limited amount of training data.

Functional Connectivity and Feature Fusion Enhance Multiclass Motor-Imagery Brain-Computer Interface Performance

(1) Background: in the field of motor-imagery brain-computer interfaces (MI-BCIs), obtaining discriminative features among multiple MI tasks poses a significant challenge. Typically, features are extracted from single electroencephalography (EEG) channels, neglecting their interconnections, which leads to limited results. To address this limitation, there has been growing interest in leveraging functional brain connectivity (FC) as a feature in MI-BCIs. However, the high inter- and intra-subject variability has so far limited its effectiveness in this domain. (2) Methods: we propose a novel signal processing framework that addresses this challenge. We extracted translation-invariant features (TIFs) obtained from a scattering convolution network (SCN) and brain connectivity features (BCFs). Through a feature fusion approach, we combined features extracted from selected channels and functional connectivity features, capitalizing on the strength of each component. Moreover, we employed a multiclass support vector machine (SVM) model to classify the extracted features. (3) Results: using a public dataset (IIa of the BCI Competition IV), we demonstrated that the feature fusion approach outperformed existing state-of-the-art methods. Notably, we found that the best results were achieved by merging TIFs with BCFs, rather than considering TIFs alone. (4) Conclusions: our proposed framework could be the key for improving the performance of a multiclass MI-BCI system.

Effect of alpha range activity on SSVEP decoding in brain-computer interfaces

Brain-computer interfaces (BCIs) facilitate direct communication between the brain and external devices. For BCI technology to be commercialized for wide scale applications, BCIs should be accurate, efficient, and exhibit consistency in performance for a wide variety of users. A core challenge is the physiological and anatomical differences amongst people, which causes a high variability amongst participants in BCI studies. Hence, it becomes necessary to analyze the mechanisms causing this variability and address them by improving the decoding algorithms. In this paper, a publicly available steady-state visual evoked potential (SSVEP) dataset is analyzed to study the effect of SSVEP flicker on the endogenous alpha power and the subsequent overall effect on the classification accuracy of the participants. It was observed that the participants with classification accuracy below 95% showed increased alpha power in their brain activities. Incorrect prediction in the decoding algorithm was observed a maximum number of times when the predicted frequency was in the range 9-12 Hz. We conclude that frequencies between 9-12 Hz may result in below par performance in some participants when canonical correlation analysis is used for classification.Clinical relevance-If alpha-band frequencies are used for SSVEP stimulation, alpha power interference in EEG may alter BCI accuracy for some users.

Can Inter-Subject Associativity Predict Data-Driven BCI Performance?

Intra- and inter-subject variability causes covariate shifts in training and testing feature spaces, resulting in low sensorimotor (SMR) brain-computer interface (BCI) performance for practical implementation. Studies involving data-driven transfer learning strategies demonstrated improving BCI performance by covariate shift adaptation. In this study, we aim to illustrate if inter-subject associativity (e.g., subjects having similar SMR brain dynamics) can predict data-driven inter-subject BCI performance. We implemented a BCI classification pipeline with a common spatial pattern, principal component analysis and linear discriminant analysis for performance evaluation. Both intra- and inter-subject BCI were evaluated in 5-Fold Validation settings. We further proposed a Bhattacharyya distance-based covariate shift score (CSS) for assessing the difference between training and testing feature domains. We performed Pearson correlation analysis to draw the relation-ship between BCI performance and CSS. Intra-subject BCI performances were significantly and negatively correlated with CSS (r = -0.94, p < 0.05). For the inter-subject experiment, BCI performances were also highly and negatively associated with CSS (r = -0.61, p < 0.05). However, this data-driven BCI evaluation framework does not necessarily manifest inter-subject associativity in BCI performance, requiring further investigations for a conclusion.Clinical relevance- If it predicts BCI performance successfully, inter-subject associativity could reduce time-consuming and annoying subject-specific calibration for the users.

Bridging the BCI illiteracy gap: a subject-to-subject semantic style transfer for EEG-based motor imagery classification

Introduction

Brain-computer interfaces (BCIs) facilitate direct interaction between the human brain and computers, enabling individuals to control external devices through cognitive processes. Despite its potential, the problem of BCI illiteracy remains one of the major challenges due to inter-subject EEG variability, which hinders many users from effectively utilizing BCI systems. In this study, we propose a subject-to-subject semantic style transfer network (SSSTN) at the feature-level to address the BCI illiteracy problem in electroencephalogram (EEG)-based motor imagery (MI) classification tasks.

Methods

Our approach uses the continuous wavelet transform method to convert high-dimensional EEG data into images as input data. The SSSTN 1) trains a classifier for each subject, 2) transfers the distribution of class discrimination styles from the source subject (the best-performing subject for the classifier, i.e., BCI expert) to each subject of the target domain (the remaining subjects except the source subject, specifically BCI illiterates) through the proposed style loss, and applies a modified content loss to preserve the class-relevant semantic information of the target domain, and 3) finally merges the classifier predictions of both source and target subject using an ensemble technique.

Results and discussion

We evaluate the proposed method on the BCI Competition IV-2a and IV-2b datasets and demonstrate improved classification performance over existing methods, especially for BCI illiterate users. The ablation experiments and t-SNE visualizations further highlight the effectiveness of the proposed method in achieving meaningful feature-level semantic style transfer.

A novel semi-supervised meta learning method for subject-transfer brain-computer interface

The brain-computer interface (BCI) provides a direct communication pathway between the human brain and external devices. However, the models trained for existing subjects perform poorly on new subjects, which is termed the subject calibration problem. In this paper, we propose a semi-supervised meta learning (SSML) method for subject-transfer calibration. The proposed SSML learns a model-agnostic meta learner with existing subjects and then fine-tunes the meta learner in a semi-supervised learning manner, i.e. using a few labelled samples and many unlabelled samples of the target subject for calibration. It is significant for BCI applications in which labelled data are scarce or expensive while unlabelled data are readily available. Three different BCI paradigms are tested: event-related potential detection, emotion recognition and sleep staging. The SSML achieved classification accuracies of 0.95, 0.89 and 0.83 in the benchmark datasets of three paradigms. The runtime complexity of SSML grows linearly as the number of samples of target subject increases so that is possible to apply it in real-time systems. This study is the first attempt to apply semi-supervised model-agnostic meta learning methodology for subject calibration. The experimental results demonstrated the effectiveness and potential of the SSML method for subject-transfer BCI applications.

Improvement of brain-computer interface in motor imagery training through the designing of a dynamic experiment and FBCSP

Motor imagery (MI) can produce a specific brain pattern when the subject imagines performing a particular action without any actual body movements. According to related previous research, the improvement of the training of MI brainwaves can be adopted by feedback methods in which the analysis of brainwave characteristics is very important. The aim of this study was to improve the subject's MI and the accuracy of classification. In order to ameliorate the accuracy of the MI of the left and right hand, the present study designed static and dynamic visual stimuli in experiments so as to evaluate which one can improve subjects' imagination training. Additionally, the filter bank common spatial pattern (FBCSP) method was used to divide the frequency band range of the brainwaves into multiple segments, following which linear discriminant analysis (LDA) was adopted for classification. The results revealed that the averaged false positive rate (FPR) under FBCSP-LDA in the dynamic MI experiment was the lowest FPR (23.76%). As such, this study suggested that a combination of the dynamic MI experiment and the FBCSP-LDA method improved the overall prediction error rate and ameliorated the performance of the MI brain-computer interface.

Discrepancy between inter- and intra-subject variability in EEG-based motor imagery brain-computer interface: Evidence from multiple perspectives

Introduction

Inter- and intra-subject variability are caused by the variability of the psychological and neurophysiological factors over time and across subjects. In the application of in Brain-Computer Interfaces (BCI), the existence of inter- and intra-subject variability reduced the generalization ability of machine learning models seriously, which further limited the use of BCI in real life. Although many transfer learning methods can compensate for the inter- and intra-subject variability to some extent, there is still a lack of clear understanding about the change of feature distribution between the cross-subject and cross-session electroencephalography (EEG) signal.

Methods

To investigate this issue, an online platform for motor-imagery BCI decoding has been built in this work. The EEG signal from both the multi-subject (Exp1) and multi-session (Exp2) experiments has been analyzed from multiple perspectives.

Results

Firstly we found that with the similar variability of classification results, the time-frequency response of the EEG signal within-subject in Exp2 is more consistent than cross-subject results in Exp1. Secondly, the standard deviation of the common spatial pattern (CSP) feature has a significant difference between Exp1 and Exp2. Thirdly, for model training, different strategies for the training sample selection should be applied for the cross-subject and cross-session tasks.

Discussion

All these findings have deepened the understanding of inter- and intra-subject variability. They can also guide practice for the new transfer learning methods development in EEG-based BCI. In addition, these results also proved that BCI inefficiency was not caused by the subject's unable to generate the event-related desynchronization/synchronization (ERD/ERS) signal during the motor imagery.

Applying correlation analysis to electrode optimization in source domain

In brain computer interface-based neurorehabilitation system, a large number of electrodes may increase the difficulty of signal acquisition and the time consumption of decoding algorithm for motor imagery EEG (MI-EEG). The traditional electrode optimization methods were limited by the low spatial resolution of scalp EEG. EEG source imaging (ESI) was further applied to reduce the number of electrodes, in which either the electrodes covering activated cortical areas were selected, or the reconstructed electrodes of EEGs with higher Fisher scores were retained. However, the activated dipoles do not all contribute equally to decoding, and the Fisher score cannot represent the correlations between electrodes and dipoles. In this paper, based on ESI and correlation analysis, a novel electrode optimization method, denoted ECCEO, was developed. The scalp MI-EEG was mapped to cortical regions by ESI, and the dipoles with larger amplitudes were chosen to designate a region of interest (ROI). Then, Pearson correlation coefficients between each dipole of the ROI and the corresponding electrode were calculated, averaged, and ranked to obtain two average correlation coefficient sequences. A small but important group of electrodes for each class were alternately added to the predetermined basic electrode set to form a candidate electrode set. Their features were extracted and evaluated to determine the optimal electrode set. Experiments were conducted on two public datasets, the average decoding accuracies achieved 95.99% and 88.30%, and the reduction of computational cost were 65% and 56%, respectively; statistical significance was examined as well.

Identifying Thematics in a Brain-Computer Interface Research

This umbrella review is motivated to understand the shift in research themes on brain-computer interfacing (BCI) and it determined that a shift away from themes that focus on medical advancement and system development to applications that included education, marketing, gaming, safety, and security has occurred. The background of this review examined aspects of BCI categorisation, neuroimaging methods, brain control signal classification, applications, and ethics. The specific area of BCI software and hardware development was not examined. A search using One Search was undertaken and 92 BCI reviews were selected for inclusion. Publication demographics indicate the average number of authors on review papers considered was 4.2 ± 1.8. The results also indicate a rapid increase in the number of BCI reviews from 2003, with only three reviews before that period, two in 1972, and one in 1996. While BCI authors were predominantly Euro-American in early reviews, this shifted to a more global authorship, which China dominated by 2020-2022. The review revealed six disciplines associated with BCI systems: life sciences and biomedicine (n = 42), neurosciences and neurology (n = 35), and rehabilitation (n = 20); (2) the second domain centred on the theme of functionality: computer science (n = 20), engineering (n = 28) and technology (n = 38). There was a thematic shift from understanding brain function and modes of interfacing BCI systems to more applied research novel areas of research-identified surround artificial intelligence, including machine learning, pre-processing, and deep learning. As BCI systems become more invasive in the lives of "normal" individuals, it is expected that there will be a refocus and thematic shift towards increased research into ethical issues and the need for legal oversight in BCI application.

A review of critical challenges in MI-BCI: From conventional to deep learning methods

Brain-computer interfaces (BCIs) have achieved significant success in controlling external devices through the Electroencephalogram (EEG) signal processing. BCI-based Motor Imagery (MI) system bridges brain and external devices as communication tools to control, for example, wheelchair for people with disabilities, robotic control, and exoskeleton control. This success largely depends on the machine learning (ML) approaches like deep learning (DL) models. DL algorithms provide effective and powerful models to analyze compact and complex EEG data optimally for MI-BCI applications. DL models with CNN network have revolutionized computer vision through end-to-end learning from raw data. Meanwhile, RNN networks have been able to decode EEG signals by processing sequences of time series data. However, many challenges in the MI-BCI field have affected the performance of DL models. A major challenge is the individual differences in the EEG signal of different subjects. Therefore, the model must be retrained from the scratch for each new subject, which leads to computational costs. Analyzing the EEG signals is challenging due to its low signal to noise ratio and non-stationary nature. Additionally, limited size of existence datasets can lead to overfitting which can be prevented by using transfer learning (TF) approaches. The main contributions of this study are discovering major challenges in the MI-BCI field by reviewing the state of art machine learning models and then suggesting solutions to address these challenges by focusing on feature selection, feature extraction and classification methods.

Department of Medicine 2040: Implementing a Constrained Disorder Principle-Based Second-Generation Artificial Intelligence System for Improved Patient Outcomes in the Department of Internal Medicine

Internal medicine departments must adapt their structures and methods of operation to accommodate changing healthcare systems. The present paper discusses some challenges departments of medicine face as healthcare providers and consumers continue to change. A co-pilot model is described in this article for augmenting physicians rather than replacing them. The paper presents the co-pilot models to improve diagnoses, treatments, and monitoring. Personalized variability patterns based on the constrained-disorder principle (CDP) are described to assess chronic therapies' effectiveness in improving patient outcomes. Based on CDP-based enhanced digital twins, this paper presents personalized treatments and follow-ups that improve diagnosis accuracy and therapy outcomes. While maintaining their professional values, departments of internal medicine must respond proactively to the needs of patients and healthcare systems. To meet the needs of patients and healthcare systems, they must strive for medical professionalism and adapt to the dynamic environment.

Wearable electroencephalography and multi-modal mental state classification: A systematic literature review

BACKGROUND

Wearable multi-modal time-series classification applications outperform their best uni-modal counterparts and hold great promise. A modality that directly measures electrical correlates from the brain is electroencephalography. Due to varying noise sources, different key brain regions, key frequency bands, and signal characteristics like non-stationarity, techniques for data pre-processing and classification algorithms are task-dependent.

METHOD

Here, a systematic literature review on mental state classification for wearable electroencephalography is presented. Four search terms in different combinations were used for an in-title search. The search was executed on the 29th of June 2022, across Google Scholar, PubMed, IEEEXplore, and ScienceDirect. 76 most relevant publications were set into context as the current state-of-the-art in mental state time-series classification.

RESULTS

Pre-processing techniques, features, and time-series classification models were analyzed. Across publications, a window length of one second was mainly chosen for classification and spectral features were utilized the most. The achieved performance per time-series classification model is analyzed, finding linear discriminant analysis, decision trees, and k-nearest neighbors models outperform support-vector machines by a factor of up to 1.5. A historical analysis depicts future trends while under-reported aspects relevant to practical applications are discussed.

CONCLUSIONS

Five main conclusions are given, covering utilization of available area for electrode placement on the head, most often or scarcely utilized features and time-series classification model architectures, baseline reporting practices, as well as explainability and interpretability of Deep Learning. The importance of a 'test battery' assessing the influence of data pre-processing and multi-modality on time-series classification performance is emphasized.

EEG-EMG coupling as a hybrid method for steering detection in car driving settings

Understanding mental processes in complex human behavior is a key issue in driving, representing a milestone for developing user-centered assistive driving devices. Here, we propose a hybrid method based on electroencephalographic (EEG) and electromyographic (EMG) signatures to distinguish left and right steering in driving scenarios. Twenty-four participants took part in the experiment consisting of recordings of 128-channel EEG and EMG activity from deltoids and forearm extensors in non-ecological and ecological steering tasks. Specifically, we identified the EEG mu rhythm modulation correlates with motor preparation of self-paced steering actions in the non-ecological task, while the concurrent EMG activity of the left (right) deltoids correlates with right (left) steering. Consequently, we exploited the mu rhythm de-synchronization resulting from the non-ecological task to detect the steering side using cross-correlation analysis with the ecological EMG signals. Results returned significant cross-correlation values showing the coupling between the non-ecological EEG feature and the muscular activity collected in ecological driving conditions. Moreover, such cross-correlation patterns discriminate the steering side earlier relative to the single EMG signal. This hybrid system overcomes the limitation of the EEG signals collected in ecological settings such as low reliability, accuracy, and adaptability, thus adding to the EMG the characteristic predictive power of the cerebral data. These results prove how it is possible to complement different physiological signals to control the level of assistance needed by the driver.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11571-021-09776-w.

A pediatric near-infrared spectroscopy brain-computer interface based on the detection of emotional valence

Brain-computer interfaces (BCIs) are being investigated as an access pathway to communication for individuals with physical disabilities, as the technology obviates the need for voluntary motor control. However, to date, minimal research has investigated the use of BCIs for children. Traditional BCI communication paradigms may be suboptimal given that children with physical disabilities may face delays in cognitive development and acquisition of literacy skills. Instead, in this study we explored emotional state as an alternative access pathway to communication. We developed a pediatric BCI to identify positive and negative emotional states from changes in hemodynamic activity of the prefrontal cortex (PFC). To train and test the BCI, 10 neurotypical children aged 8–14 underwent a series of emotion-induction trials over four experimental sessions (one offline, three online) while their brain activity was measured with functional near-infrared spectroscopy (fNIRS). Visual neurofeedback was used to assist participants in regulating their emotional states and modulating their hemodynamic activity in response to the affective stimuli. Child-specific linear discriminant classifiers were trained on cumulatively available data from previous sessions and adaptively updated throughout each session. Average online valence classification exceeded chance across participants by the last two online sessions (with 7 and 8 of the 10 participants performing better than chance, respectively, in Sessions 3 and 4). There was a small significant positive correlation with online BCI performance and age, suggesting older participants were more successful at regulating their emotional state and/or brain activity. Variability was seen across participants in regards to BCI performance, hemodynamic response, and discriminatory features and channels. Retrospective offline analyses yielded accuracies comparable to those reported in adult affective BCI studies using fNIRS. Affective fNIRS-BCIs appear to be feasible for school-aged children, but to further gauge the practical potential of this type of BCI, replication with more training sessions, larger sample sizes, and end-users with disabilities is necessary.

Decoding EEG rhythms offline and online during motor imagery for standing and sitting based on a brain-computer interface

Motor imagery (MI)-based brain-computer interface (BCI) systems have shown promising advances for lower limb motor rehabilitation. The purpose of this study was to develop an MI-based BCI for the actions of standing and sitting. Thirty-two healthy subjects participated in the study using 17 active EEG electrodes. We used a combination of the filter bank common spatial pattern (FBCSP) method and the regularized linear discriminant analysis (RLDA) technique for decoding EEG rhythms offline and online during motor imagery for standing and sitting. The offline analysis indicated the classification of motor imagery and idle state provided a mean accuracy of 88.51 ± 1.43% and 85.29 ± 1.83% for the sit-to-stand and stand-to-sit transitions, respectively. The mean accuracies of the sit-to-stand and stand-to-sit online experiments were 94.69 ± 1.29% and 96.56 ± 0.83%, respectively. From these results, we believe that the MI-based BCI may be useful to future brain-controlled standing systems.

Distribution Adaptation and Classification Framework Based on Multiple Kernel Learning for Motor Imagery BCI Illiteracy

A brain-computer interface (BCI) translates a user's thoughts such as motor imagery (MI) into the control of external devices. However, some people, who are defined as BCI illiteracy, cannot control BCI effectively. The main characteristics of BCI illiterate subjects are low classification rates and poor repeatability. To address the problem of MI-BCI illiteracy, we propose a distribution adaptation method based on multi-kernel learning to make the distribution of features between the source domain and target domain become even closer to each other, while the divisibility of categories is maximized. Inspired by the kernel trick, we adopted a multiple-kernel-based extreme learning machine to train the labeled source-domain data to find a new high-dimensional subspace that maximizes data divisibility, and then use multiple-kernel-based maximum mean discrepancy to conduct distribution adaptation to eliminate the difference in feature distribution between domains in the new subspace. In light of the high dimension of features of MI-BCI illiteracy, random forest, which can effectively handle high-dimensional features without additional cross-validation, was employed as a classifier. The proposed method was validated on an open dataset. The experimental results show that that the method we proposed suits MI-BCI illiteracy and can reduce the inter-domain differences, resulting in a reduction in the performance degradation of both cross-subjects and cross-sessions.

Closed-loop motor imagery EEG simulation for brain-computer interfaces

In a brain-computer interface (BCI) system, the testing of decoding algorithms, tasks, and their parameters is critical for optimizing performance. However, conducting human experiments can be costly and time-consuming, especially when investigating broad sets of parameters. Attempts to utilize previously collected data in offline analysis lack a co-adaptive feedback loop between the system and the user present online, limiting the applicability of the conclusions obtained to real-world uses of BCI. As such, a number of studies have attempted to address this cost-wise middle ground between offline and live experimentation with real-time neural activity simulators. We present one such system which generates motor imagery electroencephalography (EEG) via forward modeling and novel motor intention encoding models for conducting sensorimotor rhythm (SMR)-based continuous cursor control experiments in a closed-loop setting. We use the proposed simulator with 10 healthy human subjects to test the effect of three decoder and task parameters across 10 different values. Our simulated approach produces similar statistical conclusions to those produced during parallel, paired, online experimentation, but in 55% of the time. Notably, both online and simulated experimentation expressed a positive effect of cursor velocity limit on performance regardless of subject average performance, supporting the idea of relaxing constraints on cursor gain in online continuous cursor control. We demonstrate the merits of our closed-loop motor imagery EEG simulation, and provide an open-source framework to the community for closed-loop SMR-based BCI studies in the future. All code including the simulator have been made available on GitHub.

Motor Imagery Analysis from Extensive EEG Data Representations Using Convolutional Neural Networks

Motor imagery is a complex mental task that represents muscular movement without the execution of muscular action, involving cognitive processes of motor planning and sensorimotor proprioception of the body. Since the mental task has similar behavior to that of the motor execution process, it can be used to create rehabilitation routines for patients with some motor skill impairment. However, due to the nature of this mental task, its execution is complicated. Hence, the classification of these signals in scenarios such as brain-computer interface systems tends to have a poor performance. In this work, we study in depth different forms of data representation of motor imagery EEG signals for distinct CNN-based models as well as novel EEG data representations including spectrograms and multidimensional raw data. With the aid of transfer learning, we achieve results up to 93% accuracy, exceeding the current state of the art. However, although these results are strong, they entail the use of high computational resources to generate the samples, since they are based on spectrograms. Thus, we searched further for alternative forms of EEG representations, based on 1D, 2D, and 3D variations of the raw data, leading to promising results for motor imagery classification that still exceed the state of the art. Hence, in this work, we focus on exploring alternative methods to process and improve the classification of motor imagery features with few preprocessing techniques.

Classification of motor imagery EEG using deep learning increases performance in inefficient BCI users

Motor Imagery Brain-Computer Interfaces (MI-BCIs) are AI-driven systems that capture brain activity patterns associated with mental imagination of movement and convert them into commands for external devices. Traditionally, MI-BCIs operate on Machine Learning (ML) algorithms, which require extensive signal processing and feature engineering to extract changes in sensorimotor rhythms (SMR). In recent years, Deep Learning (DL) models have gained popularity for EEG classification as they provide a solution for automatic extraction of spatio-temporal features in the signals. However, past BCI studies that employed DL models, only attempted them with a small group of participants, without investigating the effectiveness of this approach for different user groups such as inefficient users. BCI inefficiency is a known and unsolved problem within BCI literature, generally defined as the inability of the user to produce the desired SMR patterns for the BCI classifier. In this study, we evaluated the effectiveness of DL models in capturing MI features particularly in the inefficient users. EEG signals from 54 subjects who performed a MI task of left- or right-hand grasp were recorded to compare the performance of two classification approaches; a ML approach vs. a DL approach. In the ML approach, Common Spatial Patterns (CSP) was used for feature extraction and then Linear Discriminant Analysis (LDA) model was employed for binary classification of the MI task. In the DL approach, a Convolutional Neural Network (CNN) model was constructed on the raw EEG signals. Additionally, subjects were divided into high vs. low performers based on their online BCI accuracy and the difference between the two classifiers’ performance was compared between groups. Our results showed that the CNN model improved the classification accuracy for all subjects within the range of 2.37 to 28.28%, but more importantly, this improvement was significantly larger for low performers. Our findings show promise for employment of DL models on raw EEG signals in future MI-BCI systems, particularly for BCI inefficient users who are unable to produce desired sensorimotor patterns for conventional ML approaches.

Decoding of Turning Intention during Walking Based on EEG Biomarkers

In the EEG literature, there is a lack of asynchronous intention models that realistically propose interfaces for applications that must operate in real time. In this work, a novel BMI approach to detect in real time the intention to turn is proposed. For this purpose, an offline, pseudo-online and online analysis is presented to validate the EEG as a biomarker for the intention to turn. This article presents a methodology for the creation of a BMI that could differentiate two classes: monotonous walk and intention to turn. A comparison of some of the most popular algorithms in the literature is conducted. To filter the signal, two relevant algorithms are used: H∞ filter and ASR. For processing and classification, the mean of the covariance matrices in the Riemannian space was calculated and then, with various classifiers of different types, the distance of the test samples to each class in the Riemannian space was estimated. This dispenses with power-based models and the necessary baseline correction, which is a problem in realistic scenarios. In the cross-validation for a generic selection (valid for any subject) and a personalized one, the results were, on average, 66.2% and 69.6% with the best filter H∞. For the pseudo-online, the custom configuration for each subject was an average of 40.2% TP and 9.3 FP/min; the best subject obtained 43.9% TP and 2.9 FP/min. In the final validation test, this subject obtained 2.5 FP/min and an accuracy rate of 71.43%, and the turn anticipation was 0.21 s on average.

Overview of the EEG-Based Classification of Motor Imagery Activities Using Machine Learning Methods and Inference Acceleration with FPGA-Based Cards

In this article, we provide a brief overview of the EEG-based classification of motor imagery activities using machine learning methods. We examined the effect of data segmentation and different neural network structures. By applying proper window size and using a purely convolutional neural network, we achieved 97.7% recognition accuracy on data from twenty subjects in three classes. The proposed architecture outperforms several networks used in previous research and makes the motor imagery-based BCI more efficient in some applications. In addition, we examined the performance of the neural network on a FPGA-based card and compared it with the inference speed and accuracy provided by a general-purpose processor.

Basal ganglia-cortical connectivity underlies self-regulation of brain oscillations in humans

Brain-computer interfaces provide an artificial link by which the brain can directly interact with the environment. To achieve fine brain-computer interface control, participants must modulate the patterns of the cortical oscillations generated from the motor and somatosensory cortices. However, it remains unclear how humans regulate cortical oscillations, the controllability of which substantially varies across individuals. Here, we performed simultaneous electroencephalography (to assess brain-computer interface control) and functional magnetic resonance imaging (to measure brain activity) in healthy participants. Self-regulation of cortical oscillations induced activity in the basal ganglia-cortical network and the neurofeedback control network. Successful self-regulation correlated with striatal activity in the basal ganglia-cortical network, through which patterns of cortical oscillations were likely modulated. Moreover, basal ganglia-cortical network and neurofeedback control network connectivity correlated with strong and weak self-regulation, respectively. The findings indicate that the basal ganglia-cortical network is important for self-regulation, the understanding of which should help advance brain-computer interface technology.

Simultaneous fMRI-EEG in 26 healthy participants indicate that the basal ganglia cortical network and the neurofeedback control network play different roles in self-regulation, providing further insight into the neural correlates for brain-machine interface control and feedback.

EEGSym: Overcoming Inter-Subject Variability in Motor Imagery Based BCIs With Deep Learning

In this study, we present a new Deep Learning (DL) architecture for Motor Imagery (MI) based Brain Computer Interfaces (BCIs) called EEGSym. Our implementation aims to improve previous state-of-the-art performances on MI classification by overcoming inter-subject variability and reducing BCI inefficiency, which has been estimated to affect 10-50% of the population. This convolutional neural network includes the use of inception modules, residual connections and a design that introduces the symmetry of the brain through the mid-sagittal plane into the network architecture. It is complemented with a data augmentation technique that improves the generalization of the model and with the use of transfer learning across different datasets. We compare EEGSym's performance on inter-subject MI classification with ShallowConvNet, DeepConvNet, EEGNet and EEG-Inception. This comparison is performed on 5 publicly available datasets that include left or right hand motor imagery of 280 subjects. This population is the largest that has been evaluated in similar studies to date. EEGSym significantly outperforms the baseline models reaching accuracies of 88.6±9.0 on Physionet, 83.3±9.3 on OpenBMI, 85.1±9.5 on Kaya2018, 87.4±8.0 on Meng2019 and 90.2±6.5 on Stieger2021. At the same time, it allows 95.7% of the tested population (268 out of 280 users) to reach BCI control (≥70% accuracy). Furthermore, these results are achieved using only 16 electrodes of the more than 60 available on some datasets. Our implementation of EEGSym, which includes new advances for EEG processing with DL, outperforms previous state-of-the-art approaches on inter-subject MI classification.

A Bayesian-optimized design for an interpretable convolutional neural network to decode and analyze the P300 response in autism

OBJECTIVE

P300 can be analyzed in autism spectrum disorder (ASD) to derive biomarkers and can be decoded in BCIs to reinforce ASD impaired skills. Convolutional neural networks (CNNs) have been proposed for P300 decoding, outperforming traditional algorithms but they i) do not investigate optimal designs in different training conditions; ii) lack in interpretability. To overcome these limitations, an interpretable CNN (ICNN), that we recently proposed for motor decoding, has been modified and adopted here, with its optimal design searched via Bayesian optimization.

APPROACH

The ICNN provides a straightforward interpretation of spectral and spatial features learned to decode P300. The Bayesian-optimized (BO) ICNN design was investigated separately for different training strategies (within-subject, within-session, and cross-subject) and BO models were used for the subsequent analyses. Specifically, transfer learning (TL) potentialities were investigated by assessing how pretrained cross-subject BO models performed on a new subject vs. random-initialized models. Furthermore, within-subject BO-derived models were combined with an Explanation Technique (ICNN+ET) to analyze P300 spectral and spatial features.

MAIN RESULTS

The ICNN resulted comparable or even outperformed existing CNNs, at the same time being lighter. Bayesian-optimized ICNN designs differed depending on the training strategy, needing more capacity as the training set variability increased. Furthermore, TL provided higher performance than networks trained from scratch. The ICNN+ET analysis suggested the frequency range [2, 5.8] Hz as the most relevant, and spatial features showed a right-hemispheric parietal asymmetry. The ICNN+ET-derived features, but not ERP-derived features, resulted significantly and highly correlated to ADOS clinical scores.

SIGNIFICANCE

This study substantiates the idea that a CNN can be designed both accurate and interpretable for P300 decoding, with an optimized design depending on the training condition. The novel ICNN-based analysis tool was able to better capture ASD neural signatures than traditional ERP analysis, possibly paving the way for identifying novel biomarkers.

How to successfully classify EEG in motor imagery BCI: a metrological analysis of the state of the art

Objective. Processing strategies are analysed with respect to the classification of electroencephalographic signals related to brain-computer interfaces based on motor imagery. A review of literature is carried out to understand the achievements in motor imagery classification, the most promising trends, and the challenges in replicating these results. Main focus is placed on performance by means of a rigorous metrological analysis carried out in compliance with the international vocabulary of metrology. Hence, classification accuracy and its uncertainty are considered, as well as repeatability and reproducibility.Approach. The paper works included in the review concern the classification of electroencephalographic signals in motor-imagery- based brain-computer interfaces. Article search was carried out in accordance with the PRISMA standard and 89 studies were included.Main results. Statistically-based analyses show that brain-inspired approaches are increasingly proposed, and that these are particularly successful in discriminating against multiple classes. Notably, many proposals involve convolutional neural networks. Instead, classical machine learning approaches are still effective for binary classifications. Many proposals combine common spatial pattern, least absolute shrinkage and selection operator, and support vector machines. Regarding reported classification accuracies, performance above the upper quartile is in the 85 % to 100 % range for the binary case and in the 83 % to 93 % range for multi-class one. Associated uncertainties are up to 6 % while repeatability for a predetermined dataset is up to 8 %. Reproducibility assessment was instead prevented by lack of standardization in experiments.Significance. By relying on the analysed studies, the reader is guided towards the development of a successful processing strategy as a crucial part of a brain-computer interface. Moreover, it is suggested that future studies should extend these approaches on data from more subjects and with custom experiments, even by investigating online operation. This would also enable the quantification of results reproducibility.