Brain-computer Interface Classification Accuracy Research Articles

Novel electrotactile brain-computer interface with somatosensory event-related potential based control.

A brain computer interface (BCI) allows users to control external devices using non-invasive brain recordings, such as electroencephalography (EEG). We developed and tested a novel electrotactile BCI prototype based on somatosensory event-related potentials (sERP) as control signals, paired with a tactile attention task as a control paradigm. A novel electrotactile BCI comprises commercial EEG device, an electrical stimulator and custom software for EEG recordings, electrical stimulation control, synchronization between devices, signal processing, feature extraction, selection, and classification. We tested a novel BCI control paradigm based on tactile attention on a sensation at a target stimulation location on the forearm. Tactile stimuli were electrical pulses delivered at two proximal locations on the user's forearm for stimulating branches of radial and median nerves, with equal probability of the target and distractor stimuli occurrence, unlike in any other ERP-based BCI design. We proposed a compact electrical stimulation electrodes configuration for delivering electrotactile stimuli (target and distractor) using 2 stimulation channels and 3 stimulation electrodes. We tested the feasibility of a single EEG channel BCI control, to determine pseudo-online BCI performance, in ten healthy subjects. For optimizing the BCI performance we compared the results for two classifiers, sERP averaging approaches, and novel dedicated feature extraction/selection methods via cross-validation procedures. We achieved a single EEG channel BCI classification accuracy in the range of 75.1 to 88.1% for all subjects. We have established an optimal combination of: single trial averaging to obtain sERP, feature extraction/selection methods and classification approach. The obtained results demonstrate that a novel electrotactile BCI paradigm with equal probability of attended (target) and unattended (distractor) stimuli and proximal stimulation sites is feasible. This method may be used to drive restorative BCIs for sensory retraining in stroke or brain injury, or assistive BCIs for communication in severely disabled users.

The performance impact of data augmentation in CSP-based motor-imagery systems for BCI applications

ObjectiveThis work investigates the performance impact of time-delay data-augmentation in motor-imagery classifiers for brain-computer-interface (BCI) applications. MethodsThe considered strategy is an extension of the Common Spectral Spatial Patterns (CSSP) method, which consists of accommodating available additional information about intra- and inter-electrode correlation into the information matrix employed by the conventional Common Spatial Patterns (CSP) method. ResultsSimulation with electroencephalographic signals obtained from two public motor-imagery datasets, in a context of differentiation between binary (left and right-hand) movements, result in an overall classification improvement of 5%, and up to 31.7% for a single individual, when compared to the use of the conventional correlation matrix in CSP-based methods. These results are supported through statistical inference and the analysis of the Receiving Operating Characteristic curve. ConclusionThe results obtained indicate that intra and inter-electrode correlation have relevant information, which have been underestimated by the literature in motor-imagery based BCI applications. SignificanceThe analyzed time-delay data-augmentation method improves the motor-imagery BCI classification accuracy with a foreseeable increase in the computation complexity.

Effects of a Vibro-Tactile P300 Based Brain-Computer Interface on the Coma Recovery Scale-Revised in Patients With Disorders of Consciousness.

Persons diagnosed with disorders of consciousness (DOC) typically suffer from motor and cognitive disabilities. Recent research has shown that non-invasive brain-computer interface (BCI) technology could help assess these patients’ cognitive functions and command following abilities. 20 DOC patients participated in the study and performed 10 vibro-tactile P300 BCI sessions over 10 days with 8–12 runs each day. Vibrotactile tactors were placed on the each patient’s left and right wrists and one foot. Patients were instructed, via earbuds, to concentrate and silently count vibrotactile pulses on either their left or right wrist that presented a target stimulus and to ignore the others. Changes of the BCI classification accuracy were investigated over the 10 days. In addition, the Coma Recovery Scale-Revised (CRS-R) score was measured before and after the 10 vibro-tactile P300 sessions. In the first run, 10 patients had a classification accuracy above chance level (>12.5%). In the best run, every patient reached an accuracy ≥60%. The grand average accuracy in the first session for all patients was 40%. In the best session, the grand average accuracy was 88% and the median accuracy across all sessions was 21%. The CRS-R scores compared before and after 10 VT3 sessions for all 20 patients, are showing significant improvement (p = 0.024). Twelve of the twenty patients showed an improvement of 1 to 7 points in the CRS-R score after the VT3 BCI sessions (mean: 2.6). Six patients did not show a change of the CRS-R and two patients showed a decline in the score by 1 point. Every patient achieved at least 60% accuracy at least once, which indicates successful command following. This shows the importance of repeated measures when DOC patients are assessed. The improvement of the CRS-R score after the 10 VT3 sessions is an important issue for future experiments to test the possible therapeutic applications of vibro-tactile and related BCIs with a larger patient group.

An Optimized Channel Selection Method Based on Multifrequency CSP-Rank for Motor Imagery-Based BCI System.

Background Due to the redundant information contained in multichannel electroencephalogram (EEG) signals, the classification accuracy of brain-computer interface (BCI) systems may deteriorate to a large extent. Channel selection methods can help to remove task-independent electroencephalogram (EEG) signals and hence improve the performance of BCI systems. However, in different frequency bands, brain areas associated with motor imagery are not exactly the same, which will result in the inability of traditional channel selection methods to extract effective EEG features. New Method To address the above problem, this paper proposes a novel method based on common spatial pattern- (CSP-) rank channel selection for multifrequency band EEG (CSP-R-MF). It combines the multiband signal decomposition filtering and the CSP-rank channel selection methods to select significant channels, and then linear discriminant analysis (LDA) was used to calculate the classification accuracy. Results The results showed that our proposed CSP-R-MF method could significantly improve the average classification accuracy compared with the CSP-rank channel selection method.

Investigation of a wavelet-based neural network learning algorithm applied to P300 based brain-computer interface

The paper presented herein proposes an algorithm that aims at improving the classification accuracy of Brain-Computer Interface (BCI) speller. In this work, feed-forward neural network with back propagation learning is used for classification purposes. Testing of the proposed algorithm was performed through the utilization of two datasets, namely; Berlin BCI Competition III and EPFL BCI groups. Results, for the first dataset, indicated that the use of 64 electrodes with 30 hidden layers grants an accuracy of 94.9 %, while an average accuracy of 95.8% (range: 92%-100%) was obtained for the second dataset when using a 32 electrode configuration with 20 hidden layers. The obtained accuracy levels, in this study, are higher when compared with other recent classification approaches.

Sinusoidal Signal Assisted Multivariate Empirical Mode Decomposition for Brain-Computer Interfaces.

A brain-computer interface (BCI) is a communication approach that permits cerebral activity to control computers or external devices. Brain electrical activity recorded with electroencephalography (EEG) is most commonly used for BCI. Noise-assisted multivariate empirical mode decomposition (NA-MEMD) is a data-driven time-frequency analysis method that can be applied to nonlinear and nonstationary EEG signals for BCI data processing. However, because white Gaussian noise occupies a broad range of frequencies, some redundant components are introduced. To solve this leakage problem, in this study, we propose using a sinusoidal assisted signal that occupies the same frequency ranges as the original signals to improve MEMD performance. To verify the effectiveness of the proposed sinusoidal signal assisted MEMD (SA-MEMD) method, we compared the decomposition performances of MEMD, NA-MEMD, and the proposed SA-MEMD using synthetic signals and a real-world BCI dataset. The spectral decomposition results indicate that the proposed SA-MEMD can avoid the generation of redundant components and over decomposition, thus, substantially reduce the mode mixing and misalignment that occurs in MEMD and NA-MEMD. Moreover, using SA-MEMD as a signal preprocessing method instead of MEMD or NA-MEMD can significantly improve BCI classification accuracy and reduce calculation time, which indicates that SA-MEMD is a powerful spectral decomposition method for BCI.

Localization of brain electrical activity sources and hemodynamic activity foci during motor imagery

Studied are sources of brain activity contributing to EEG patterns which correspond to motor imagery. The accuracy of their classification determines the efficiency of brain-computer interface (BCI) allowing for controlling external technical devices directly by brain signals without involving muscle activity. Sources of brain activity are identified by Independent Component Analysis. Those independent components for which the BCI classification accuracy are at maximum are treated as relevant for motor imagery task. Two of the most relevant sources demonstrate strictly exposed event related desynchronization and synchronization of mu--rhythm during imagery of contra--and ipsilateral hands. These sources are localized by solving inverse EEG problem taking into account individual geometry of brain and its covers provided by anatomical MRI images. The sources are shown to be localized in BA 3A relating to proprioceptive sensitivity of the contralateral hand. Their positions are closed to foci of BOLD activity obtained by fMRI.

Performance of Motor Imagery Brain-Computer Interface Based on Anodal Transcranial Direct Current Stimulation Modulation

Voluntarily modulating neural activity plays a key role in brain-computer interface (BCI). In general, the self-regulated neural activation patterns are used in the current BCI systems involving the repetitive trainings with feedback for an attempt to achieve a high-quality control performance. With the limitation posed by the training procedure in most BCI studies, the present work aims to investigate whether directly modulating the neural activity by using an external method could facilitate the BCI control. We designed an experimental paradigm that combines anodal transcranial direct current stimulation (tDCS) with a motor imagery (MI)-based feedback EEG BCI system. Thirty-two young and healthy human subjects were randomly assigned to the real and sham stimulation groups to evaluate the effect of tDCS-induced EEG pattern changes on BCI classification accuracy. Results showed that the anodal tDCS obviously induces sensorimotor rhythm (SMR)-related event-related desynchronization (ERD) pattern changes in the upper-mu (10-14 Hz) and beta (14-26 Hz) rhythm components. Both the online and offline BCI classification results demonstrate that the enhancing ERD patterns could conditionally improve BCI performance. This pilot study suggests that the tDCS is a promising method to help the users to develop reliable BCI control strategy in a relatively short time.

Improving classification accuracy using intra-session classifier training and implementation for a BCI based on automated parameter selection

Genetic Algorithms (GAs) were used in a previous study to automate parameter selection for an EEG-based P300-driven Brain-Computer Interface (BCI). The GA approach showed marked improvement over data-insensitive parameter selection; however, it required lengthy execution times thereby rendering it infeasible for online implementation. Automated parameter selection is retained in this work; however, it is achieved using the less computationally intensive N-fold cross-validation (NFCV). Additionally, this study sought to improve BCI classification accuracy using a training data collection and application protocol that the authors refer to as 'Intra-session classifier training and implementation'. Intra-session classifier training and implementation using NFCV-driven automated parameter selection yielded a classification accuracy of 82.94% compared to 45.44% for the inter-session approach using data-insensitive parameters. These findings are significant impact since the intra-session protocol can be applied to any P300-based BCI regardless of its application platform to obtain improved classification accuracy.

Walking by Thinking: The Brainwaves Are Crucial, Not the Muscles!

Healthy participants are able to move forward within a virtual environment (VE) by the imagination of foot movement. This is achieved by using a brain-computer interface (BCI) that transforms thought-modulated electroencephalogram (EEG) recordings into a control signal. A BCI establishes a communication channel between the human brain and the computer. The basic principle of the Graz-BCI is the detection and classification of motor-imagery-related EEG patterns, whereby the dynamics of sensorimotor rhythms are analyzed. A BCI is a closed-loop system and information is visually fed back to the user about the success or failure of an intended movement imagination. Feedback can be realized in different ways, from a simple moving bar graph to navigation in VEs. The goals of this work are twofold: first, to show the influence of different feedback types on the same task, and second, to demonstrate that it is possible to move through a VE (e.g., a virtual street) without any muscular activity, using only the imagination of foot movement. In the presented work, data from BCI feedback displayed on a conventional monitor are compared with data from BCI feedback in VE experiments with a head-mounted display (HMD) and in a high immersive projection environment (Cave). Results of three participants are reported to demonstrate the proof-of-concept. The data indicate that the type of feedback has an influence on the task performance, but not on the BCI classification accuracy. The participants achieved their best performances viewing feedback in the Cave. Furthermore the VE feedback provided motivation for the subjects.

Exploring Virtual Environments with an EEG-based BCI through Motor Imagery / Erkundung von virtuellen Welten durch Bewegungsvorstellungen mit Hilfe eines EEG-basierten BCI

In this paper, we describe the possibility of navigating in a virtual environment using the output signal of an EEG-based Brain-Computer Interface (BCI). The graphical capabilities of virtual reality (VR) should help to create new BCI-paradigms and improve feedback presentation. The objective of this combination is to enhance the subject's learning process of gaining control of the BCI. In this study, the participant had to imagine left or right hand movements while exploring a virtual conference room. By imaging a left hand movement the subject turned virtually to the left inside the room and with right hand imagery to the right. In fact, three trained subjects reached 80% to 100% BCI classification accuracy in the course of the experimental sessions. All subjects were able to achieve a rotation in the VR to the left or right by approximately 45 degrees during one trial.