Unraveling motor imagery brain patterns using explainable artificial intelligence based on Shapley values.

BACKGROUND AND OBJECTIVE: Motor imagery (MI) based brain-computer interfaces (BCIs) are widely used in rehabilitation due to the close relationship that exists between MI and motor execution (ME). However, the underlying brain mechanisms of MI remain not well understood. Most MI-BCIs use the sensorimotor rhythms elicited in the primary motor cortex (M1) and somatosensory cortex (S1), which consist of an event-related desynchronization followed by an event-related synchronization. Consequently, this has resulted in systems that only record signals around M1 and S1. However, MI could involve a more complex network including sensory, association, and motor areas. In this study, we hypothesize that the superior accuracies achieved by new deep learning (DL) models applied to MI decoding rely on focusing on a broader MI activation of the brain. Parallel to the success of DL, the field of explainable artificial intelligence (XAI) has seen continuous development to provide explanations for DL networks success. The goal of this study is to use XAI in combination with DL to extract information about MI brain activation patterns from non-invasive electroencephalography (EEG) signals. METHODS: We applied an adaptation of Shapley additive explanations (SHAP) to EEGSym, a state-of-the-art DL network with exceptional transfer learning capabilities for inter-subject MI classification. We obtained the SHAP values from two public databases comprising 171 users generating left and right hand MI instances with and without real-time feedback. RESULTS: We found that EEGSym based most of its prediction on the signal of the frontal electrodes, i.e. F7 and F8, and on the first 1500 ms of the analyzed imagination period. We also found that MI involves a broad network not only based on M1 and S1, but also on the prefrontal cortex (PFC) and the posterior parietal cortex (PPC). We further applied this knowledge to select a 8-electrode configuration that reached inter-subject accuracies of 86.5% ± 10.6% on the Physionet dataset and 88.7% ± 7.0% on the Carnegie Mellon University's dataset. CONCLUSION: Our results demonstrate the potential of combining DL and SHAP-based XAI to unravel the brain network involved in producing MI. Furthermore, SHAP values can optimize the requirements for out-of-laboratory BCI applications involving real users.

Embodied strategies for public speaking anxiety: evaluation of the Corp-Oral program.

Background: Public speaking is an indispensable skill that can profoundly influence success in both professional and personal spheres. Regrettably, managing anxiety during a speech poses a significant challenge for many of the population. This research assessed the impacts of a Corp-Oral program, designed to manage public speaking anxiety in university students, based on, body awareness, embodied message techniques, simulation, embodied visualization, body transformation, and gesture enhancement. Methods: Thirty-six students (61% women; Mage = 20.22, SD = 1.23 years) were randomly assigned to either an experimental group (n = 18), which underwent the Corp-Oral program, or a control group (n = 18). Self-perceived anxiety, heart rate, and electroencephalography were measured in a pre-test and a post-test. Results: The study reveals that the Corp-Oral program significantly (p < 0.005) reduced both physiological responses (heart rate) and self-reported measures of anxiety. The alteration was more noticeable in self-reported anxiety measures (a decrease of 33.217%) than in heart rate (a decrease of 4.659%). During the speech, the experimental group exhibited increased cortical activation in areas related to emotional regulation, consciousness, sensorimotor integration, and movement control. A significant increase in frontal alpha asymmetry was observed for the experimental group in the post-test, but there were no significant variations in the theta/beta ratio. Conclusion: These findings underline the benefit of managing public speaking anxiety not merely by reducing it but by channeling it through embodied strategies. These strategies could lead to greater action awareness that would cushion the physiological effect of the anxiety response and help generate a better self-perception of the anxiety state.

Assessing the Potential of Brain-Computer Interface Multiplayer Video Games using c-VEPs: A Pilot Study.

Video games have become a common and widespread form of entertainment, while non-invasive brain-computer interfaces (BCI) are emerging as potential alternative communication technologies. Combining BCIs and video games can enhance the gaming experience and make it accessible to motor-disabled individuals. Recently, code-modulated visual evoked potentials (c-VEP) have been proposed as a novel control signal able to achieve high performance with short calibration times. However, there are still no video games that use c-VEPs as a control signal. The aim of this pilot study is to develop an implementation of the 'Connect 4' multiplayer video game using a c-VEP-based BCI and test it with 10 healthy users. Participants were paired to compete in matches and carried out individual tasks. The results showed that the participants were able to control the game with an average accuracy of 94.10% and a selection time of 5.25 seconds per command, outperforming previous approaches. This suggests that the proposed video game is feasible and c-VEPs can provide smooth BCI control.

Nonparametric Early Stopping Detection for c-VEP-based Brain-Computer Interfaces: A Pilot Study.

Brain-computer interface (BCI) systems based on code-modulated visual evoked potentials (c-VEP) stand out for achieving excellent command selection accuracies with very short calibration times. One of the natural steps to democratize their use in plug-and-play environments is to develop early stopping algorithms. These methods allow real-time detection of the minimum number of code repetitions needed to provide reliable selections. However, such techniques are scarce in the current state-of-the-art for c-VEP-based BCI systems based on the classical circular shifting paradigm. Here, a novel nonparametric early stopping method is proposed, which approximates the distribution of unattended commands to a normal distribution and issues a selection when the correlation of the command is considered an outlier. The proposal has been evaluated offline with 15 healthy users, achieving an average accuracy of 97.08% and a speed of 1.37 s/command. Likewise, the algorithm has also been evaluated with an additional user in an online way, as a proof of concept to validate its technical feasibility, achieving an average accuracy of 96.88% with a speed of 1.67 s/command. These results suggest that the real time application of the proposed algorithm is feasible, significantly reducing the required selection time without compromising accuracy.

Influence of spatial frequency in visual stimuli for cVEP-based BCIs: evaluation of performance and user experience.

Code-modulated visual evoked potentials (c-VEPs) are an innovative control signal utilized in brain-computer interfaces (BCIs) with promising performance. Prior studies on steady-state visual evoked potentials (SSVEPs) have indicated that the spatial frequency of checkerboard-like stimuli influences both performance and user experience. Spatial frequency refers to the dimensions of the individual squares comprising the visual stimulus, quantified in cycles (i.e., number of black-white squares pairs) per degree of visual angle. However, the specific effects of this parameter on c-VEP-based BCIs remain unexplored. Therefore, the objective of this study is to investigate the role of spatial frequency of checkerboard-like visual stimuli in a c-VEP-based BCI. Sixteen participants evaluated selection matrices with eight spatial frequencies: C001 (0 c/°, 1×1 squares), C002 (0.15 c/°, 2×2 squares), C004 (0.3 c/°, 4×4 squares), C008 (0.6 c/°, 8×8 squares), C016 (1.2 c/°, 16×16 squares), C032 (2.4 c/°, 32×32 squares), C064 (4.79 c/°, 64×64 squares), and C128 (9.58 c/°, 128×128 squares). These conditions were tested in an online spelling task, which consisted of 18 trials each conducted on a 3×3 command interface. In addition to accuracy and information transfer rate (ITR), subjective measures regarding comfort, ocular irritation, and satisfaction were collected. Significant differences in performance and comfort were observed based on different stimulus spatial frequencies. Although all conditions achieved mean accuracy over 95% after 2.1 s of trial duration, C016 stood out in terms user experience. The proposed condition not only achieved a mean accuracy of 96.53% and 164.54 bits/min with a trial duration of 1.05s, but also was reported to be significantly more comfortable than the traditional C001 stimulus. Since both features are key for BCI development, higher spatial frequencies than the classical black-to-white stimulus might be more adequate for c-VEP systems. Hence, we assert that the spatial frequency should be carefully considered in the development of future applications for c-VEP-based BCIs.

Combining brain-computer interfaces and multiplayer video games: an application based on c-VEPs.

Introduction and objective: Video games are crucial to the entertainment industry, nonetheless they can be challenging to access for those with severe motor disabilities. Brain-computer interfaces (BCI) systems have the potential to help these individuals by allowing them to control video games using their brain signals. Furthermore, multiplayer BCI-based video games may provide valuable insights into how competitiveness or motivation affects the control of these interfaces. Despite the recent advancement in the development of code-modulated visual evoked potentials (c-VEPs) as control signals for high-performance BCIs, to the best of our knowledge, no studies have been conducted to develop a BCI-driven video game utilizing c-VEPs. However, c-VEPs could enhance user experience as an alternative method. Thus, the main goal of this work was to design, develop, and evaluate a version of the well-known 'Connect 4' video game using a c-VEP-based BCI, allowing 2 users to compete by aligning 4 same-colored coins vertically, horizontally or diagonally. Methods: The proposed application consists of a multiplayer video game controlled by a real-time BCI system processing 2 electroencephalograms (EEGs) sequentially. To detect user intention, columns in which the coin can be placed was encoded with shifted versions of a pseudorandom binary code, following a traditional circular shifting c-VEP paradigm. To analyze the usability of our application, the experimental protocol comprised an evaluation session by 22 healthy users. Firstly, each user had to perform individual tasks. Afterward, users were matched and the application was used in competitive mode. This was done to assess the accuracy and speed of selection. On the other hand, qualitative data on satisfaction and usability were collected through questionnaires. Results: The average accuracy achieved was 93.74% ± 1.71%, using 5.25 seconds per selection. The questionnaires showed that users felt a minimal workload. Likewise, high satisfaction values were obtained, highlighting that the application was intuitive and responds quickly and smoothly. Conclusions: This c-VEP based multiplayer video game has reached suitable performance on 22 users, supported by high motivation and minimal workload. Consequently, compared to other versions of "Connect 4" that utilized different control signals, this version has exhibited superior performance.

ITACA: An open-source framework for Neurofeedback based on Brain-Computer Interfaces.

BACKGROUND AND OBJECTIVE: Neurofeedback (NF) is a paradigm that allows users to self-modulate patterns of brain activity. It is implemented with a closed-loop brain-computer interface (BCI) system that analyzes the user's brain activity in real-time and provides continuous feedback. This paradigm is of great interest due to its potential as a non-pharmacological and non-invasive alternative to treat non-degenerative brain disorders. Nevertheless, currently available NF frameworks have several limitations, such as the lack of a wide variety of real-time analysis metrics or overly simple training scenarios that may negatively affect user performance. To overcome these limitations, this work proposes ITACA: a novel open-source framework for the design, implementation and evaluation of NF training paradigms. METHODS: ITACA is designed to be easy-to-use, flexible and attractive. Specifically, ITACA includes three different gamified training scenarios with a choice of five brain activity metrics as real-time feedback. Among them, novel metrics based on functional connectivity and network theory stand out. It is complemented with five different computerized versions of widespread cognitive assessment tests. To validate the proposed framework, a computational efficiency analysis and an NF training protocol focused on frontal-medial theta modulation were conducted. RESULTS: Efficiency analysis proved that all implemented metrics allow an optimal feedback update rate for conducting NF sessions. Furthermore, conducted NF protocol yielded results that support the use of ITACA in NF research studies. CONCLUSIONS: ITACA implements a wide variety of features for designing, conducting and evaluating NF studies with the goal of helping researchers expand the current state-of-the-art in NF training.

MEDUSA©: A novel Python-based software ecosystem to accelerate brain-computer interface and cognitive neuroscience research.

BACKGROUND AND OBJECTIVE: Neurotechnologies have great potential to transform our society in ways that are yet to be uncovered. The rate of development in this field has increased significantly in recent years, but there are still barriers that need to be overcome before bringing neurotechnologies to the general public. One of these barriers is the difficulty of performing experiments that require complex software, such as brain-computer interfaces (BCI) or cognitive neuroscience experiments. Current platforms have limitations in terms of functionality and flexibility to meet the needs of researchers, who often need to implement new experimentation settings. This work was aimed to propose a novel software ecosystem, called MEDUSA©, to overcome these limitations. METHODS: We followed strict development practices to optimize MEDUSA© for research in BCI and cognitive neuroscience, making special emphasis in the modularity, flexibility and scalability of our solution. Moreover, it was implemented in Python, an open-source programming language that reduces the development cost by taking advantage from its high-level syntax and large number of community packages. RESULTS: MEDUSA© provides a complete suite of signal processing functions, including several deep learning architectures or connectivity analysis, and ready-to-use BCI and neuroscience experiments, making it one of the most complete solutions nowadays. We also put special effort in providing tools to facilitate the development of custom experiments, which can be easily shared with the community through an app market available in our website to promote reproducibility. CONCLUSIONS: MEDUSA© is a novel software ecosystem for modern BCI and neurotechnology experimentation that provides state-of-the-art tools and encourages the participation of the community to make a difference for the progress of these fields. Visit the official website at https://www.medusabci.com/ to know more about this project.

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.

Robust asynchronous control of ERP-Based brain-Computer interfaces using deep learning.

BACKGROUND AND OBJECTIVE: Brain-computer interfaces (BCI) based on event-related potentials (ERP) are a promising technology for alternative and augmented communication in an assistive context. However, most approaches to date are synchronous, requiring the intervention of a supervisor when the user wishes to turn his attention away from the BCI system. In order to bring these BCIs into real-life applications, a robust asynchronous control of the system is required through monitoring of user attention. Despite the great importance of this limitation, which prevents the deployment of these systems outside the laboratory, it is often overlooked in research articles. This study was aimed to propose a novel method to solve this problem, taking advantage of deep learning for the first time in this context to overcome the limitations of previous strategies based on hand-crafted features. METHODS: The proposed method, based on EEG-Inception, a novel deep convolutional neural network, divides the problem in 2 stages to achieve the asynchronous control: (i) the model detects user's control state, and (ii) decodes the command only if the user is attending to the stimuli. Additionally, we used transfer learning to reduce the calibration time, even exploring a calibration-less approach. RESULTS: Our method was evaluated with 22 healthy subjects, analyzing the impact of the calibration time and number of stimulation sequences on the system's performance. For the control state detection stage, we report average accuracies above 91% using only 1 sequence of stimulation and 30 calibration trials, reaching a maximum of 96.95% with 15 sequences. Moreover, our calibration-less approach also achieved suitable results, with a maximum accuracy of 89.36%, showing the benefits of transfer learning. As for the overall asynchronous system, which includes both stages, the maximum information transfer rate was 35.54 bpm, a suitable value for high-speed communication. CONCLUSIONS: The proposed strategy achieved higher performance with less calibration trials and stimulation sequences than former approaches, representing a promising step forward that paves the way for more practical applications of ERP-based spellers.

Neurofeedback Training Based on Motor Imagery Strategies Increases EEG Complexity in Elderly Population.

Neurofeedback training (NFT) has shown promising results in recent years as a tool to address the effects of age-related cognitive decline in the elderly. Since previous studies have linked reduced complexity of electroencephalography (EEG) signal to the process of cognitive decline, we propose the use of non-linear methods to characterise changes in EEG complexity induced by NFT. In this study, we analyse the pre- and post-training EEG from 11 elderly subjects who performed an NFT based on motor imagery (MI-NFT). Spectral changes were studied using relative power (RP) from classical frequency bands (delta, theta, alpha, and beta), whilst multiscale entropy (MSE) was applied to assess EEG-induced complexity changes. Furthermore, we analysed the subject's scores from Luria tests performed before and after MI-NFT. We found that MI-NFT induced a power shift towards rapid frequencies, as well as an increase of EEG complexity in all channels, except for C3. These improvements were most evident in frontal channels. Moreover, results from cognitive tests showed significant enhancement in intellectual and memory functions. Therefore, our findings suggest the usefulness of MI-NFT to improve cognitive functions in the elderly and encourage future studies to use MSE as a metric to characterise EEG changes induced by MI-NFT.

Brain-computer interfaces based on code-modulated visual evoked potentials (c-VEP): a literature review.

Objective.Code-modulated visual evoked potentials (c-VEP) have been consolidated in recent years as robust control signals capable of providing non-invasive brain-computer interfaces (BCIs) for reliable, high-speed communication. Their usefulness for communication and control purposes has been reflected in an exponential increase of related articles in the last decade. The aim of this review is to provide a comprehensive overview of the literature to gain understanding of the existing research on c-VEP-based BCIs, since its inception (1984) until today (2021), as well as to identify promising future research lines.Approach.The literature review was conducted according to the Preferred Reporting Items for Systematic reviews and Meta-Analysis guidelines. After assessing the eligibility of journal manuscripts, conferences, book chapters and non-indexed documents, a total of 70 studies were included. A comprehensive analysis of the main characteristics and design choices of c-VEP-based BCIs was discussed, including stimulation paradigms, signal processing, modeling responses, applications, etc.Main results.The literature review showed that state-of-the-art c-VEP-based BCIs are able to provide an accurate control of the system with a large number of commands, high selection speeds and even without calibration. In general, a lack of validation in real setups was observed, especially regarding the validation with disabled populations. Future work should be focused toward developing self-paced c-VEP-based portable BCIs applied in real-world environments that could exploit the unique benefits of c-VEP paradigms. Some aspects such as asynchrony, unsupervised training, or code optimization still require further research and development.Significance.Despite the growing popularity of c-VEP-based BCIs, to the best of our knowledge, this is the first literature review on the topic. In addition to providing a joint discussion of the advances in the field, some future lines of research are suggested to contribute to the development of reliable plug-and-play c-VEP-based BCIs.

A new method to build multiplex networks using canonical correlation analysis for the characterization of the Alzheimer's disease continuum.

Objective. The aim of this study was to solve one of the current limitations for the characterization of the brain network in the Alzheimer's disease (AD) continuum. Nowadays, frequency-dependent approaches have reached contradictory results depending on the frequency band under study, tangling the possible clinical interpretations.Approach. To overcome this issue, we proposed a new method to build multiplex networks based on canonical correlation analysis (CCA). Our method determines two basis vectors using the source and electrode-level frequency-specific network parameters for a reference group, and then project the results for the rest of the groups into these hyperplanes to make them comparable. It was applied to: (i) synthetic signals generated with a Kuramoto-based model; and (ii) a resting-state electroencephalography (EEG) database formed by recordings from 51 cognitively healthy controls, 51 mild cognitive impairment subjects, 51 mild AD patients, 50 moderate AD patients, and 50 severe AD patients.Main results. Our results using synthetic signals showed that the interpretation of the proposed CCA-based multiplex parameters (multiplex strength, multiplex characteristic path length and multiplex clustering coefficient) can be analogous to their frequency-specific counterparts, as they displayed similar behaviors in terms of average connectivity, integration, and segregation. Findings using real EEG recordings revealed that dementia due to AD is characterized by a significant increase in average connectivity, and by a loss of integration and segregation.Significance. We can conclude that CCA can be used to build multiplex networks based from frequency-specific results, summarizing all the available information and avoiding the limitations of possible frequency-specific conflicts. Additionally, our method supposes a novel approach for the construction and analysis of multiplex networks during AD continuum.

A Convolutional Neural Network Architecture to Enhance Oximetry Ability to Diagnose Pediatric Obstructive Sleep Apnea.

This study aims at assessing the usefulness of deep learning to enhance the diagnostic ability of oximetry in the context of automated detection of pediatric obstructive sleep apnea (OSA). A total of 3196 blood oxygen saturation (SpO2) signals from children were used for this purpose. A convolutional neural network (CNN) architecture was trained using 20-min SpO2 segments from the training set (859 subjects) to estimate the number of apneic events. CNN hyperparameters were tuned using Bayesian optimization in the validation set (1402 subjects). This model was applied to three test sets composed of 312, 392, and 231 subjects from three independent databases, in which the apnea-hypopnea index (AHI) estimated for each subject (AHICNN) was obtained by aggregating the output of the CNN for each 20-min SpO2 segment. AHICNN outperformed the 3% oxygen desaturation index (ODI3), a clinical approach, as well as the AHI estimated by a conventional feature-engineering approach based on multi-layer perceptron (AHIMLP). Specifically, AHICNN reached higher four-class Cohen's kappa in the three test databases than ODI3 (0.515 vs 0.417, 0.422 vs 0.372, and 0.423 vs 0.369) and AHIMLP (0.515 vs 0.377, 0.422 vs 0.381, and 0.423 vs 0.306). In addition, our proposal outperformed state-of-the-art studies, particularly for the AHI severity cutoffs of 5 e/h and 10 e/h. This suggests that the information automatically learned from the SpO2 signal by deep-learning techniques helps to enhance the diagnostic ability of oximetry in the context of pediatric OSA.

Correction: Martínez-Cagigal, V.; Santamaría-Vázquez, E.; Hornero, R. Asynchronous Control of P300-Based Brain-Computer Interfaces Using Sample Entropy. Entropy 2019, 21, 230.

Figure 5 of the original paper contains errors [...].

EEG-Inception: A Novel Deep Convolutional Neural Network for Assistive ERP-Based Brain-Computer Interfaces.

In recent years, deep-learning models gained attention for electroencephalography (EEG) classification tasks due to their excellent performance and ability to extract complex features from raw data. In particular, convolutional neural networks (CNN) showed adequate results in brain-computer interfaces (BCI) based on different control signals, including event-related potentials (ERP). In this study, we propose a novel CNN, called EEG-Inception, that improves the accuracy and calibration time of assistive ERP-based BCIs. To the best of our knowledge, EEG-Inception is the first model to integrate Inception modules for ERP detection, which combined efficiently with other structures in a light architecture, improved the performance of our approach. The model was validated in a population of 73 subjects, of which 31 present motor disabilities. Results show that EEG-Inception outperforms 5 previous approaches, yielding significant improvements for command decoding accuracy up to 16.0%, 10.7%, 7.2%, 5.7% and 5.1% in comparison to rLDA, xDAWN + Riemannian geometry, CNN-BLSTM, DeepConvNet and EEGNet, respectively. Moreover, EEG-Inception requires very few calibration trials to achieve state-of-the-art performances taking advantage of a novel training strategy that combines cross-subject transfer learning and fine-tuning to increase the feasibility of this approach for practical use in assistive applications.

BCIAUT-P300: A Multi-Session and Multi-Subject Benchmark Dataset on Autism for P300-Based Brain-Computer-Interfaces.

There is a lack of multi-session P300 datasets for Brain-Computer Interfaces (BCI). Publicly available datasets are usually limited by small number of participants with few BCI sessions. In this sense, the lack of large, comprehensive datasets with various individuals and multiple sessions has limited advances in the development of more effective data processing and analysis methods for BCI systems. This is particularly evident to explore the feasibility of deep learning methods that require large datasets. Here we present the BCIAUT-P300 dataset, containing 15 autism spectrum disorder individuals undergoing 7 sessions of P300-based BCI joint-attention training, for a total of 105 sessions. The dataset was used for the 2019 IFMBE Scientific Challenge organized during MEDICON 2019 where, in two phases, teams from all over the world tried to achieve the best possible object-detection accuracy based on the P300 signals. This paper presents the characteristics of the dataset and the approaches followed by the 9 finalist teams during the competition. The winner obtained an average accuracy of 92.3% with a convolutional neural network based on EEGNet. The dataset is now publicly released and stands as a benchmark for future P300-based BCI algorithms based on multiple session data.

Asynchronous Control of ERP-Based BCI Spellers Using Steady-State Visual Evoked Potentials Elicited by Peripheral Stimuli.

Brain-computer interface (BCI) spellers based on event related potentials (ERPs) are intrinsically synchronous systems. Therefore, selections are constantly made, even when users are not paying attention to the stimuli. This poses a major limitation in real-life applications, in which an asynchronous control is required. The aim of this study is to design, develop and test a novel method to discriminate whether the user is controlling the system (i.e., control state) or is engaged in other task (i.e., non-control state). To achieve such an asynchronous control, our method detects the steady-state visual evoked potentials (SSVEPs) elicited by peripheral stimuli of ERP-based spellers. A characterization experiment was conducted with 5 subjects to investigate general aspects of this phenomenon. Then, the proposed method was validated with 15 subjects in offline and online sessions. Results show that the proposed method provides a reliable asynchronous control, achieving an average accuracy of 95.5% for control state detection during the online sessions. Furthermore, our approach is independent of the ERP classification stage, and to the best of our knowledge, is the first procedure that does not need to extend the duration of the calibration sessions to acquire non-control observations.

Towards Automatic Artifact Rejection in Resting-State MEG Recordings: Evaluating the Performance of the SOUND Algorithm.

In this study, a new automated noise rejection algorithm, the SOurce-estimate-Utilizing Noise-Discarding algorithm (SOUND), was evaluated on magnetoencephalographic (MEG) resting-state signals in order to select its optimal configuration parameters. Different values of the epoch length and the regularization parameter λ0 were assessed in three scenarios with ascending noise levels. Results show that it is possible to remarkably improve the Signal-to-Noise Ratio, without overly altering the signal of interest. An optimal λ0 value of 0.1 was obtained. However, the epoch length should be adapted to the specific problem. In conclusion, our results suggest that the SOUND algorithm is an appropriate and useful tool to be applied in a preprocessing pipeline for MEG restingstate signals.

Asynchronous Control of P300-Based Brain-Computer Interfaces Using Sample Entropy.

Brain-computer interfaces (BCI) have traditionally worked using synchronous paradigms. In recent years, much effort has been put into reaching asynchronous management, providing users with the ability to decide when a command should be selected. However, to the best of our knowledge, entropy metrics have not yet been explored. The present study has a twofold purpose: (i) to characterize both control and non-control states by examining the regularity of electroencephalography (EEG) signals; and (ii) to assess the efficacy of a scaled version of the sample entropy algorithm to provide asynchronous control for BCI systems. Ten healthy subjects participated in the study, who were asked to spell words through a visual oddball-based paradigm, attending (i.e., control) and ignoring (i.e., non-control) the stimuli. An optimization stage was performed for determining a common combination of hyperparameters for all subjects. Afterwards, these values were used to discern between both states using a linear classifier. Results show that control signals are more complex and irregular than non-control ones, reaching an average accuracy of 94 . 40 % in classification. In conclusion, the present study demonstrates that the proposed framework is useful in monitoring the attention of a user, and granting the asynchrony of the BCI system.