FBDNN: filter banks and deep neural networks for portable and fast brain-computer interfaces.

Objective.To propose novel SSVEP classification methodologies using deep neural networks (DNNs) and improve performances in single-channel and user-independent brain-computer interfaces (BCIs) with small data lengths.Approach.We propose the utilization of filter banks (creating sub-band components of the EEG signal) in conjunction with DNNs. In this context, we created three different models: a recurrent neural network (FBRNN) analyzing the time domain, a 2D convolutional neural network (FBCNN-2D) processing complex spectrum features and a 3D convolutional neural network (FBCNN-3D) analyzing complex spectrograms, which we introduce in this study as possible input for SSVEP classification. We tested our neural networks on three open datasets and conceived them so as not to require calibration from the final user, simulating a user-independent BCI.Results.The DNNs with the filter banks surpassed the accuracy of similar networks without this preprocessing step by considerable margins, and they outperformed common SSVEP classification methods (SVM and FBCCA) by even higher margins.Conclusion and significance.Filter banks allow different types of deep neural networks to more efficiently analyze the harmonic components of SSVEP. Complex spectrograms carry more information than complex spectrum features and the magnitude spectrum, allowing the FBCNN-3D to surpass the other CNNs. The performances obtained in the challenging classification problems indicates a strong potential for the construction of portable, economical, fast and low-latency BCIs.

Motor imagery practice and feedback effects on functional connectivity.

- Objective: the use of motor imagery (MI) in motor rehabilitation protocols has been increasingly investigated as a potential technique for enhancing traditional treatments, yielding better clinical outcomes. However, since MI performance can be challenging, practice is usually required. This demands appropriate training, actively engaging the MI-related brain areas, consequently enabling the user to properly benefit from it. The role of feedback is central for MI practice. Yet, assessing which underlying neural changes are feedback-specific or purely due to MI practice is still a challenging effort, mainly due to the difficulty in isolating their contributions. In this work, we aimed to assess functional connectivity (FC) changes following MI practice that are either extrinsic or specific to feedback. APPROACH: to achieve this, we investigated FC, using graph theory, in electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) data, during MI performance and at resting-state (rs), respectively. Thirty healthy subjects were divided into three groups, receiving no feedback (control), "false" feedback (sham) or actual neurofeedback (active). Participants underwent 12 to 13 hands-MI EEG sessions and pre- and post-MI training fMRI exams. MAIN RESULTS: following MI practice, control participants presented significant increases in degree and in eigenvector centrality for occipital nodes at rs-fMRI scans, whereas sham-feedback produced similar effects, but to a lesser extent. Therefore, MI practice, by itself, seems to stimulate visual information processing mechanisms that become apparent during basal brain activity. Additionally, only the active group displayed decreases in inter-subject FC patterns, both during MI performance and at rs-fMRI. SIGNIFICANCE: hence, actual neurofeedback impacted FC by disrupting common inter-subject patterns, suggesting that subject-specific neural plasticity mechanisms become important. Future studies should consider this when designing experimental NFBT protocols and analyses.

Space-time filter for SSVEP brain-computer interface based on the minimum variance distortionless response.

Brain-computer interfaces (BCI) based on steady-state visually evoked potentials (SSVEP) have been increasingly used in different applications, ranging from entertainment to rehabilitation. Filtering techniques are crucial to detect the SSVEP response since they can increase the accuracy of the system. Here, we present an analysis of a space-time filter based on the Minimum Variance Distortionless Response (MVDR). We have compared the performance of a BCI-SSVEP using the MVDR filter to other classical approaches: Common Average Reference (CAR) and Canonical Correlation Analysis (CCA). Moreover, we combined the CAR and MVDR techniques, totalling four filtering scenarios. Feature extraction was performed using Welch periodogram, Fast Fourier transform, and CCA (as extractor) with one and two harmonics. Feature selection was performed by forward wrappers, and a linear classifier was employed for discrimination. The main analyses were carried out over a database of ten volunteers, considering two cases: four and six visual stimuli. The results show that the BCI-SSVEP using the MVDR filter achieves the best performance among the analysed scenarios. Interestingly, the system's accuracy using the MVDR filter is practically constant even when the number of visual stimuli was increased, whereas degradation was observed for the other techniques.

Channel capacity in brain-computer interfaces.

OBJECTIVE: Adapted from the concept of channel capacity, the information transfer rate (ITR) has been widely used to evaluate the performance of a brain-computer interface (BCI). However, its traditional formula considers the model of a discrete memoryless channel in which the transition matrix presents very particular symmetries. As an alternative to compute the ITR, this work indicates a more general closed-form expression-also based on that channel model, but with less restrictive assumptions-and, with the aid of a selection heuristic based on a wrapper algorithm, extends such formula to detect classes that deteriorate the operation of a BCI system. APPROACH: The benchmark is a steady-state visually evoked potential (SSVEP)-based BCI dataset with 40 frequencies/classes, in which two scenarios are tested: (1) our proposed formula is used and the classes are gradually evaluated in the order of the class labels provided with the dataset; and (2) the same formula is used but with the classes evaluated progressively by a wrapper algorithm. In both scenarios, the canonical correlation analysis (CCA) is the tool to detect SSVEPs. MAIN RESULTS: Before and after class selection using this alternative ITR, the average capacity among all subjects goes from 3.71 [Formula: see text] 1.68 to 4.79 [Formula: see text] 0.70 bits per symbol, with p -value <0.01, and, for a supposedly BCI-illiterate subject, her/his capacity goes from 1.53 to 3.90 bits per symbol. SIGNIFICANCE: Besides indicating a consistent formula to compute ITR, this work provides an efficient method to perform channel assessment in the context of a BCI experiment and argues that such method can be used to study BCI illiteracy.

Space-time recurrences for functional connectivity evaluation and feature extraction in motor imagery brain-computer interfaces.

This work presents a classification performance comparison between different frameworks for functional connectivity evaluation and complex network feature extraction aiming to distinguish motor imagery classes in electroencephalography (EEG)-based brain-computer interfaces (BCIs). The analysis was performed in two online datasets: (1) a classical benchmark-the BCI competition IV dataset 2a-allowing a comparison with a representative set of strategies previously employed in this BCI paradigm and (2) a statistically representative dataset for signal processing technique comparisons over 52 subjects. Besides exploring three classical similarity measures-Pearson correlation, Spearman correlation, and mean phase coherence-this work also proposes a recurrence-based alternative for estimating EEG brain functional connectivity, which takes into account the recurrence density between pairwise electrodes over a time window. These strategies were followed by graph feature evaluation considering clustering coefficient, degree, betweenness centrality, and eigenvector centrality. The features were selected by Fisher's discriminating ratio and classification was performed by a least squares classifier in agreement with classical and online BCI processing strategies. The results revealed that the recurrence-based approach for functional connectivity evaluation was significantly better than the other frameworks, which is probably associated with the use of higher order statistics underlying the electrode joint probability estimation and a higher capability of capturing nonlinear inter-relations. There were no significant differences in performance among the evaluated graph features, but the eigenvector centrality was the best feature regarding processing time. Finally, the best ranked graph-based attributes were found in classical EEG motor cortex positions for the subjects with best performances, relating functional organization and motor activity. Graphical Abstract Evaluating functional connectivity based on Space-Time Recurrence Counting for motor imagery classification in brain-computer interfaces. Recurrences are evaluated between electrodes over a time window, and, after a density threshold, the electrodes adjacency matrix is stablish, leading to a graph. Graph-based topological measures are used for motor imagery classification.

Analysis of User Interaction with a Brain-Computer Interface Based on Steady-State Visually Evoked Potentials: Case Study of a Game.

This paper presents a systematic analysis of a game controlled by a Brain-Computer Interface (BCI) based on Steady-State Visually Evoked Potentials (SSVEP). The objective is to understand BCI systems from the Human-Computer Interface (HCI) point of view, by observing how the users interact with the game and evaluating how the interface elements influence the system performance. The interactions of 30 volunteers with our computer game, named "Get Coins," through a BCI based on SSVEP, have generated a database of brain signals and the corresponding responses to a questionnaire about various perceptual parameters, such as visual stimulation, acoustic feedback, background music, visual contrast, and visual fatigue. Each one of the volunteers played one match using the keyboard and four matches using the BCI, for comparison. In all matches using the BCI, the volunteers achieved the goals of the game. Eight of them achieved a perfect score in at least one of the four matches, showing the feasibility of the direct communication between the brain and the computer. Despite this successful experiment, adaptations and improvements should be implemented to make this innovative technology accessible to the end user.

EEG sensorimotor rhythms' variation and functional connectivity measures during motor imagery: linear relations and classification approaches.

Hands motor imagery (MI) has been reported to alter synchronization patterns amongst neurons, yielding variations in the mu and beta bands' power spectral density (PSD) of the electroencephalography (EEG) signal. These alterations have been used in the field of brain-computer interfaces (BCI), in an attempt to assign distinct MI tasks to commands of such a system. Recent studies have highlighted that information may be missing if knowledge about brain functional connectivity is not considered. In this work, we modeled the brain as a graph in which each EEG electrode represents a node. Our goal was to understand if there exists any linear correlation between variations in the synchronization patterns-that is, variations in the PSD of mu and beta bands-induced by MI and alterations in the corresponding functional networks. Moreover, we (1) explored the feasibility of using functional connectivity parameters as features for a classifier in the context of an MI-BCI; (2) investigated three different types of feature selection (FS) techniques; and (3) compared our approach to a more traditional method using the signal PSD as classifier inputs. Ten healthy subjects participated in this study. We observed significant correlations (p < 0.05) with values ranging from 0.4 to 0.9 between PSD variations and functional network alterations for some electrodes, prominently in the beta band. The PSD method performed better for data classification, with mean accuracies of (90 ± 8)% and (87 ± 7)% for the mu and beta band, respectively, versus (83 ± 8)% and (83 ± 7)% for the same bands for the graph method. Moreover, the number of features for the graph method was considerably larger. However, results for both methods were relatively close, and even overlapped when the uncertainties of the accuracy rates were considered. Further investigation regarding a careful exploration of other graph metrics may provide better alternatives.

Automatic repositioning of jaw segments for three-dimensional virtual treatment planning of orthognathic surgery.

PURPOSE: To develop a computer-based method for automating the repositioning of jaw segments in the skull during three-dimensional virtual treatment planning of orthognathic surgery. The method speeds up the planning phase of the orthognathic procedure, releasing surgeons from laborious and time-consuming tasks. MATERIALS AND METHODS: The method finds the optimal positions for the maxilla, mandibular body, and bony chin in the skull. Minimization of cephalometric differences between measured and standard values is considered. Cone-beam computed tomographic images acquired from four preoperative patients with skeletal malocclusion were used for evaluating the method. RESULTS: Dentofacial problems of the four patients were rectified, including skeletal malocclusion, facial asymmetry, and jaw discrepancies. CONCLUSIONS: The results show that the method is potentially able to be used in routine clinical practice as support for treatment-planning decisions in orthognathic surgery.

Unorganized machines for seasonal streamflow series forecasting.

Modern unorganized machines--extreme learning machines and echo state networks--provide an elegant balance between processing capability and mathematical simplicity, circumventing the difficulties associated with the conventional training approaches of feedforward/recurrent neural networks (FNNs/RNNs). This work performs a detailed investigation of the applicability of unorganized architectures to the problem of seasonal streamflow series forecasting, considering scenarios associated with four Brazilian hydroelectric plants and four distinct prediction horizons. Experimental results indicate the pertinence of these models to the focused task.

Characterization of multiscroll attractors using Lyapunov exponents and Lagrangian coherent structures.

The present work aims to apply a recently proposed method for estimating Lyapunov exponents to characterize-with the aid of the metric entropy and the fractal dimension-the degree of information and the topological structure associated with multiscroll attractors. In particular, the employed methodology offers the possibility of obtaining the whole Lyapunov spectrum directly from the state equations without employing any linearization procedure or time series-based analysis. As a main result, the predictability and the complexity associated with the phase trajectory were quantified as the number of scrolls are progressively increased for a particular piecewise linear model. In general, it is shown here that the trajectory tends to increase its complexity and unpredictability following an exponential behaviour with the addition of scrolls towards to an upper bound limit, except for some degenerated situations where a non-uniform grid of scrolls is attained. Moreover, the approach employed here also provides an easy way for estimating the finite time Lyapunov exponents of the dynamics and, consequently, the Lagrangian coherent structures for the vector field. These structures are particularly important to understand the stretching/folding behaviour underlying the chaotic multiscroll structure and can provide a better insight of phase space partition and exploration as new scrolls are progressively added to the attractor.

An extended echo state network using Volterra filtering and principal component analysis.

Echo state networks (ESNs) can be interpreted as promoting an encouraging compromise between two seemingly conflicting objectives: (i) simplicity of the resulting mathematical model and (ii) capability to express a wide range of nonlinear dynamics. By imposing fixed weights to the recurrent connections, the echo state approach avoids the well-known difficulties faced by recurrent neural network training strategies, but still preserves, to a certain extent, the potential of the underlying structure due to the existence of feedback loops within the dynamical reservoir. Moreover, the overall training process is relatively simple, as it amounts essentially to adapting the readout, which usually corresponds to a linear combiner. However, the linear nature of the output layer may limit the capability of exploring the available information, since higher-order statistics of the signals are not taken into account. In this work, we present a novel architecture for an ESN in which the linear combiner is replaced by a Volterra filter structure. Additionally, the principal component analysis technique is used to reduce the number of effective signals transmitted to the output layer. This idea not only improves the processing capability of the network, but also preserves the simplicity of the training process. The proposed architecture is then analyzed in the context of a set of representative information extraction problems, more specifically supervised and unsupervised channel equalization, and blind separation of convolutive mixtures. The obtained results, when compared to those produced by already proposed ESN versions, highlight the benefits brought by the novel network proposal and characterize it as a promising tool to deal with challenging signal processing tasks.