Spatio-Temporal EEG Models for Brain Interfaces

Introducing block-Toeplitz covariance matrices to remaster linear discriminant analysis for event-related potential brain-computer interfaces

Objective.Covariance matrices of noisy multichannel electroencephalogram (EEG) time series data provide essential information for the decoding of brain signals using machine learning methods. However, small datasets and high dimensionality make it hard to estimate these matrices. In brain-computer interfaces (BCI) based on event-related potentials (ERP) and a linear discriminant analysis (LDA) classifier, the state of the art covariance estimation uses shrinkage regularization. As this is a general covariance regularization approach, we aim at improving LDA further by better exploiting the domain-specific characteristics of the EEG to regularize the covariance estimates.Approach.We propose to enforce a block-Toeplitz structure for the covariance matrix of the LDA, which implements an assumption of signal stationarity in short time windows for each channel.Main results.An offline re-analysis of data collected from 213 subjects under 13 different event-related potential BCI protocols showed a significantly increased binary classification performance of this 'ToeplitzLDA' compared to shrinkage regularized LDA (up to 6 AUC points,p < 0.001) and Riemannian classification approaches (up to 2 AUC points,p < 0.001). In an unsupervised visual speller application, this improvement would translate to a relative reduction of spelling errors by 81% on average for 25 subjects. Additionally, aside from lower memory and reduced time complexity for LDA training, ToeplitzLDA proves to be robust against drastic increases of the number of temporal features.Significance.The proposed covariance estimation allows BCI researchers to improve classification rates and reduce calibration times of BCI protocols using event-related potentials and thus support the usability of corresponding applications. Its lower computational and memory needs could make it a valuable algorithm especially for mobile BCIs.

Classification of Event-Related Potentials with Regularized Spatiotemporal LCMV Beamforming

The usability of EEG-based visual brain–computer interfaces (BCIs) based on event-related potentials (ERPs) benefits from reducing the calibration time before BCI operation. Linear decoding models, such as the spatiotemporal beamformer model, yield state-of-the-art accuracy. Although the training time of this model is generally low, it can require a substantial amount of training data to reach functional performance. Hence, BCI calibration sessions should be sufficiently long to provide enough training data. This work introduces two regularized estimators for the beamformer weights. The first estimator uses cross-validated L2-regularization. The second estimator exploits prior information about the structure of the EEG by assuming Kronecker–Toeplitz-structured covariance. The performances of these estimators are validated and compared with the original spatiotemporal beamformer and a Riemannian-geometry-based decoder using a BCI dataset with P300-paradigm recordings for 21 subjects. Our results show that the introduced estimators are well-conditioned in the presence of limited training data and improve ERP classification accuracy for unseen data. Additionally, we show that structured regularization results in lower training times and memory usage, and a more interpretable classification model.

Feedback Related Potentials for EEG-Based Typing Systems

Error related potentials (ErrP), which are elicited in the EEG in response to a perceived error, have been used for error correction and adaption in the event related potential (ERP)-based brain computer interfaces designed for typing. In these typing interfaces, ERP evidence is collected in response to a sequence of stimuli presented usually in the visual form and the intended user stimulus is probabilistically inferred (stimulus with highest probability) and presented to the user as the decision. If the inferred stimulus is incorrect, ErrP is expected to be elicited in the EEG. Early approaches to use ErrP in the design of typing interfaces attempt to make hard decisions on the perceived error such that the perceived error is corrected and either the sequence of stimuli are repeated to obtain further ERP evidence, or without further repetition the stimulus with the second highest probability is presented to the user as the decision of the system. Moreover, none of the existing approaches use a language model to increase the performance of typing. In this work, unlike the existing approaches, we study the potential benefits of fusing feedback related potentials (FRP), a form of ErrP, with ERP and context information (language model, LM) in a Bayesian fashion to detect the user intent. We present experimental results based on data from 12 healthy participants using RSVP Keyboard™ to complete a copy-phrase-task. Three paradigms are compared: [P1] uses only ERP/LM Bayesian fusion; [P2] each RSVP sequence is appended with the top candidate in the alphabet according to posterior after ERP evidence fusion; corresponding FRP is then incorporated; and [P3] the top candidate is shown as a prospect to generate FRP evidence only if its posterior exceeds a threshold. Analyses indicate that ERP/LM/FRP evidence fusion during decision making yields significant speed-accuracy benefits for the user.

Double-Criteria Active Learning for Multiclass Brain-Computer Interfaces

Recent technological advances have enabled researchers to collect large amounts of electroencephalography (EEG) signals in labeled and unlabeled datasets. It is expensive and time consuming to collect labeled EEG data for use in brain-computer interface (BCI) systems, however. In this paper, a novel active learning method is proposed to minimize the amount of labeled, subject-specific EEG data required for effective classifier training, by combining measures of uncertainty and representativeness within an extreme learning machine (ELM). Following this approach, an ELM classifier was first used to select a relatively large batch of unlabeled examples, whose uncertainty was measured through the best-versus-second-best (BvSB) strategy. The diversity of each sample was then measured between the limited labeled training data and previously selected unlabeled samples, and similarity is measured among the previously selected samples. Finally, a tradeoff parameter is introduced to control the balance between informative and representative samples, and these samples are then used to construct a powerful ELM classifier. Extensive experiments were conducted using benchmark and multiclass motor imagery EEG datasets to evaluate the efficacy of the proposed method. Experimental results show that the performance of the new algorithm exceeds or matches those of several state-of-the-art active learning algorithms. It is thereby shown that the proposed method improves classifier performance and reduces the need for training samples in BCI applications.

Adversarial Deep Learning in EEG Biometrics

Deep learning methods for person identification based on electroencephalographic (EEG) brain activity encounters the problem of exploiting the temporally correlated structures or recording session specific variability within EEG. Furthermore, recent methods have mostly trained and evaluated based on single session EEG data. We address this problem from an invariant representation learning perspective. We propose an adversarial inference approach to extend such deep learning models to learn session-invariant person-discriminative representations that can provide robustness in terms of longitudinal usability. Using adversarial learning within a deep convolutional network, we empirically assess and show improvements with our approach based on longitudinally collected EEG data for person identification from half-second EEG epochs.

An Event-Driven AR-Process Model for EEG-Based BCIs With Rapid Trial Sequences

Electroencephalography (EEG) is an effective non-invasive measurement method to infer user intent in brain-computer interface (BCI) systems for control and communication, however, these systems often lack sufficient accuracy and speed due to low separability of class-conditional EEG feature distributions. Many factors impact system performance, including inadequate training datasets and models' ignorance of the temporal dependency of brain responses to serial stimuli. Here, we propose a signal model for event-related responses in the EEG evoked with a rapid sequence of stimuli in BCI applications. The model describes the EEG as a superposition of impulse responses time-locked to stimuli corrupted with an autoregressive noise process. The performance of the signal model is assessed in the context of RSVP keyboard, a language-model-assisted EEG-based BCI for typing. EEG data obtained for model calibration from 10 healthy participants are used to fit and compare two models: the proposed sequence-based EEG model and the trial-based feature-class-conditional distribution model that ignores temporal dependencies, which has been used in the previous work. The simulation studies indicate that the earlier model that ignores temporal dependencies may be causing drastic reductions in achievable information transfer rate (ITR). Furthermore, the proposed model, with better regularization, may achieve improved accuracy with fewer calibration data samples, potentially helping to reduce calibration time. Specifically, results show an average 8.6% increase in (cross-validated) calibration AUC for a single channel of EEG, and 54% increase in the ITR in a typing task.

A Parametric EEG Signal Model for BCIs with Rapid-Trial Sequences

Electroencephalogram (EEG) signals have been shown very effective for inferring user intents in brain-computer interface (BCI) applications. However, existing EEG-based BCIs, in many cases, lack sufficient performance due to utilizing classifiers that operate on EEG signals induced by individual trials. While many factors influence the classification performance, an important aspect that is often ignored is the temporal dependency of these trial-EEG signals, in some cases impacted by interference of brain responses to consecutive target and non-target trials. In this study, the EEG signals are analyzed in a parametric sequence-based fashion, which considers all trials that induce brain responses in a rapid-sequence fashion, including a mixture of consecutive target and non-target trials. EEG signals are described as a linear combination of time-shifted cortical source activities plus measurement noise. Using a superposition of time invariant with an auto-regressive (AR) process, EEG signals are treated as a linear combination of a stationary Gaussian process and time-locked impulse responses to the stimulus (input events) onsets. The model performance is assessed in the framework of a rapid serial visualization presentation (RSVP) based typing task for three healthy subjects across two sessions. Signal modeling in this fashion yields promising performance outcomes considering a single EEG channel to estimate the user intent.