ICLabel is an important plug-in function in EEGLAB, the most widely used software for EEG data processing. A powerful approach to automated processing of EEG data involves decomposing the data by Independent Component Analysis (ICA) and then classifying the resulting independent components (ICs) using ICLabel. While EEGLAB pipelines support high-performance computing (HPC) platforms running the open-source Octave interpreter, the ICLabel plug-in is incompatible with Octave because of its specialized neural network architecture. To enhance cross-platform compatibility, we developed a Python version of ICLabel that uses standard EEGLAB data structures. We compared ICLabel MATLAB and Python implementations to data from 14 subjects. ICLabel returns the likelihood of classification in 7 classes of components for each ICA component. The returned IC classifications were virtually identical between Python and MATLAB, with differences in classification percentage below 0.001%.

EEG continues to find a multitude of uses in both neuroscience research and medical practice, and independent component analysis (ICA) continues to be an important tool for analyzing EEG. A multitude of ICA algorithms for EEG decomposition exist, and in the past, their relative effectiveness has been studied. AMICA is considered the benchmark against which to compare the performance of other ICA algorithms for EEG decomposition. AMICA exposes many parameters to the user to allow for precise control of the decomposition. However, several of the parameters currently tend to be set according to “rules of thumb” shared in the EEG community. Here, 70-channel AMICA decompositions are run on data from a collection of participants while varying certain key parameters. The running time and quality of decompositions are analyzed based on two metrics, Pairwise Mutual Information (PMI) and Mutual Information Reduction (MIR), and derived recommendations for selecting parameter values are presented.

Independent component analysis (ICA), is a blind source separation method that is becoming increasingly used to separate brain and non-brain related activities in electroencephalographic (EEG) and other electrophysiological recordings. It can be used to extract effective brain source activities and estimate their cortical source areas, and is commonly used in machine learning applications to classify EEG artifacts. Previously, we compared results of decomposing 1371-channel scalp EEG datasets using 22 ICA and other blind source separation (BSS) algorithms. We are now making this framework available to the scientific community and, in the process of its release are testing a recent ICA algorithm (Picard) not included in the previous assay. Our test framework uses three main metrics to assess BSS performance: Pairwise Mutual Information (PMI) between scalp channel pairs; PMI remaining between component pairs after decomposition; and, the complete (not pairwise) Mutual Information Reduction (MIR) produced by each algorithm. We also measure the “dipolarity” of the scalp projection maps for the decomposed component, defined by the number of components whose scalp projection maps nearly match the projection of a single equivalent dipole located in the volume of a template boundary element method (BEM) electrical forward problem head model. Within this framework, Picard performed similarly to Infomax ICA. This is not surprising since Picard is a type of Infomax algorithm that uses the LBFGS method for faster convergence, in contrast to Infomax and Extended Infomax (runica) which use gradient descent. Our results show that Picard performs similarly to Infomax and, likewise, better than other BSS algorithms, excepting the more computationally complex AMICA. Further research might determine if partial Picard decomposition, followed by AMICA, might produce unequaled performance without a large time penalty. We have released the source code of our framework and the test data through GitHub to encourage further comparisons of ICA/BSS algorithm performance applied to electrophysiological data (https://github.con/sccn/testica).

Convolutional Neural Networks (CNNs) have achieved impressive performance on many computer vision-related tasks, such as object detection, image recognition, image retrieval, etc. These achievements benefit from the CNNs’ outstanding capability to learn discriminative features with deep layers of neuron structures and iterative training processes. This has inspired the EEG research community to adopt CNN in performing EEG classification tasks. However, CNNs learned features are not immediately interpretable, causing a lack of understanding of the CNNs’ internal working mechanism. To improve CNN interpretability, CNN visualization methods are applied to translate the internal features into visually perceptible patterns for qualitative analysis of CNN layers. Many CNN visualization methods have been proposed in the Computer Vision literature to interpret the CNN network structure, operation, and semantic concept, yet applications to EEG data analysis have been limited. In this work we use 3 different methods to extract EEG-relevant features from a CNN trained on raw EEG data: optimal samples for each classification category, activation maximization, and reverse convolution. We applied these methods to a high-performing Deep Learning model with state-of-the-art performance for an EEG sex classification task, and show that the model exploits differences between classes in the theta frequency band. We show that the visualization of a CNN model can reveal interesting EEG biomarkers. Using these tools, EEG researchers using Deep Learning can better identify the learned EEG features, possibly identifying new class-relevant biomarkers.

Here, we introduce a novel approach to estimating noise covariance matrices for scalp projection maps of ICA-decomposed EEG sources, and show that they are useful for estimating cortical EEG source distributions. To determine spatial uncertainty characteristics of individual independent component (IC) maps returned by the AMICA decomposition [1], we used the RELICA (Bootstrap-ICA) toolbox [2] to generate 50 decompositions of bootstrap resampled versions of the same EEG data set. This allows identification of clusters of near-identical bootstrap ICs of independent component (IC) maps, each matching the scalp map of a localizable brain effective source in the full data set (reference) AMICA decomposition. This, in turn, makes it possible to estimate the spatial variability of the associated whole-data IC scalp map. For cortical source localization we used the Sparse Compact Smooth (SCS) algorithm of Cao applied to an electrical forward problem head model optimized with a SCALE estimate of individual skull conductivity. When component scalp map noise covariance matrix used in SCS was initialized to the RELICA-derived covariance map (rather than ignored), we observed an improvement in residual variance left unexplained by SCS source localization to a compact cortical patch (or pair of patches). In addition, the peak of the estimated source patch moved, by an average of 14.6 mm (range: 2-20 mm) and in some cases was localized to a different sulcus or gyrus.

Phase-Amplitude Coupling (PAC) in electrophysiological signals refers to the transient interplay of activities in different frequency ranges, wherein phase in a low-frequency band and amplitude in a high-frequency band are in some way dependent. PAC phenomena have received increasing interest in neuroscience given the growing evidence of their apparent role in both normal and pathological brain processes. This interest has resulted in publication of a wave of methods for PAC estimation, each with its own advantages and drawbacks relative to others. Motivated by the widespread study of this phenomenon, most academic open source software environments for analyzing electrophysiological signals (e.g., Fieldtrip, Brainstorm, MNE) have implemented at least one method of PAC estimation. Here we describe an EEGLAB plug-in release, PACTools for MATLAB (The Mathworks, Inc.), that computes PAC in either continuous or event-related data using any of five methods for PAC estimation: Mean Vector Length Modulation Index, Kullback- Leibler Modulation Index, Phase-Locking Value, General Linear Model Modulation Index and Mutual Information PAC (MIPAC). PACTools uses parallelized code for efficient performance and offers built-in direct access to online high-performance computing resources made freely available for nonprofit research by the Neurosciences Gateway (nsgportal.org). PACTools features intuitive graphic user interfaces and equivalent command line calls that make its use straightforward and seamless in the EEGLAB environment. We discuss toolbox implementation, architecture, dependencies, and the implemented methods of PAC computation and visualization.

Performance of classifiers on electrophysiological signals are often affected by volume conduction, thus compromising their reliability and classification accuracy. This issue is usually incorrectly overlooked when dealing with electrocorticography (ECoG) recordings. Here we propose that preprocessing ECoG signals using Independent Component Analysis (ICA) can improve classification performance. To test this hypothesis we use ECoG signals measured from the cortical surface of an epileptic subject. ECoG signals from subtemporal cortex were recorded while a series of face and house images were displayed briefly. We compare the performance of house versus face classifiers using features extracted from the recorded signals versus their independent components (ICs). We show that classification accuracy based on IC features is preserved when the channels with the highest single channel classification accuracy are removed from the analysis. Hence, features of independent signal spaces derived by ICA decomposition may improve the robustness and reliability of signal-based Brain-Computer Interface (BCI) classifiers.

The past decade has seen a gradual expansion of Brain-Computer Interface applications from their clinical roots into entertainment, automotive, workplace, and military domains. However, many of these new-found applications have yet to pass the prototype stage, among others due to challenges posed by real-world data noise levels and increased context variability and complexity. Tackling these challenges requires sufficiently realistic and rich datasets that allow for benchmarking competing approaches, and in this paper, we present a task battery modeled after a complex real-world scenario, together with a new open dataset for BCI research. Results from an exemplary analysis of a slice of this large trove of data are presented, and it is found that the data mirror some of the challenges encountered in real-world deployments, with various well-known BCI algorithms showing a pronounced performance differential to highly simplified lab experiments. The results also show significant performance differences among alternative methods, indicating possible trajectories for future improvements BCI methodology applied to complex real-world contexts.

A promising application of brain machine interfaces (BMIs) is predicting user cognitive state, particularly in complex and demanding scenarios, so that automation can dynamically and adaptively adjust task parameters to optimize joint human-machine performance. In this paper we analyze neural, physiological and behavioral data recorded during a complex two-person “crew station” task and investigate whether these measures provide information for inferring user decision state. Specifically, we investigate how measures of EEG, pupil dilation, heart rate and response time, can be fused to infer decision confidence and accuracy in two side-tasks occurring throughout a three hour experimental session. One side-task is an auditory task, the other a visual task, both occurring within the context of the crew station scenario (auditory alert and a visual satellite map N-back task). We find that the best prediction performance always fuses EEG and pupil dilation measures, with results yielding between 70%–75% accuracy with respect to whether the subject(s) will skip making the decision (i.e. have high uncertainty) or whether he/she makes an error. Interestingly, the results suggest a possible mechanistic explanation for the utility of the fused measures, specifically the interaction between the locus coeruleus (LC), whose activity is linked to arousal state and can be inferred from pupil dilation, and the anterior cingulate (ACC), which has been linked to decision formation and monitoring and whose activity is typically measured via EEG. In general, our results demonstrate the potential in using fused neuro/physio measures to infer and track human operator decision uncertainty during demanding complex tasks, possibly enabling BMIs to eventually be employed as “cognitive orthotics” for improving man-machine interaction and performance.

We analyze the jointly-sparse signal recovery problem in the regime where the number of sources k is larger than the number of measurements M. We show that the support set of sources can still be recovered with sparse Bayesian learning (M-SBL) even if k ≥ M. We provide sufficient conditions on the dictionary and sources which theoretically guarantee support set recovery in the noiseless case of M-SBL. We validate our sufficient conditions with experiments and also demonstrate that M-SBL outperforms M-CoSaMP, the algorithm recently used to localize more sources than sensors. Finally, we experimentally show robustness of the approach in the presence of noise.

Here we present an analysis of a 12-subject electroencephalographic (EEG) data set in which participants were asked to engage in prolonged, self-paced episodes of guided emotion imagination with eyes closed. Our goal is to correctly predict, given a short EEG segment, whether the participant was imagining a positive respectively negative-valence emotional scenario during the given segment using a predictive model learned via machine learning. The challenge lies in generalizing to novel (i.e., previously unseen) emotion episodes from a wide variety of scenarios including love, awe, frustration, anger, etc. based purely on spontaneous oscillatory EEG activity without stimulus event-locked responses. Using a variant of the Filter-Bank Common Spatial Pattern algorithm, we achieve an average accuracy of 71.3% correct classification of binary valence rating across 12 different emotional imagery scenarios under rigorous block-wise cross-validation.

A non-invasive way to monitor a music listener's level of engagement could give us a valuable tool for music classification, technology, and therapy. To investigate whether musical engagement can be monitored, we developed an experimental protocol using the mobile brain/body imaging (MoBI) paradigm in which participants make expressive rhythmic arm gestures to encourage and/or index musical engagement. Participants communicate the feeling pulse of music they are hearing via simple rhythmic U-shaped back-and-forth hand/arm 'conducting' gesture cycles that animate, in real time, the mirroring movement of a spot of light on a video display in front of them. Participants are asked to imagine that this display is also being viewed remotely by a deaf friend to whom they are attempting to communicate the feeling of the music they are hearing. In an Engaged condition, listeners are encouraged to fully engage themselves in this musical/emotional communication task. In a Not Engaged condition, a concurrent internal arithmetic distractor task is introduced to induce less fully engaged listening. Here, we report results of training a classifier using a frequency-based common spatial patterns (FBCSP) approach to correctly distinguish Engaged and Not Engaged conditions from concurrently recorded EEG data. Here the approach gave 67% classification accuracy across subjects (versus 50% chance), and 85% accuracy within subjects, cross-validated using a block wise paradigm.

We propose a method to solve the multiple measurement vector (MMV) sparse signal recovery problem in a robust manner when data contains outlier points which do not fit the shared sparsity structure otherwise contained in the data. This scenario occurs frequently in the applications of MMV models due to only partially known source dynamics. The algorithm we propose is a modification of MMV-based sparse bayesian learning (M-SBL) by incorporating the idea of least trimmed squares (LTS), which has previously been developed for robust linear regression. Experiments show a significant performance improvement over the conventional M-SBL under different outlier ratios and amplitudes.

This study proposes and evaluates a recursive algorithm for incremental estimation of independent components from on-line data. The algorithm offers the convergence properties of batch independent component analysis (ICA) with incremental updates of a form similar to natural gradient (NG) on-line information maximization (Infomax). We employ recursive procedure to arrive at steady state solution given by NG Infomax. Furthermore, we propose a novel procedure to compute corrective updates on the basis of previous estimates. Implementation of this algorithm incurs linear complexity in data size, input dimensions, and number of estimated independent components. Significant gains in convergence rate over on-line natural gradient ICA are demonstrated.

Because of the increasing portability and wearability of noninvasive electrophysiological systems that record and process electrical signals from the human brain, automated systems for assessing changes in user cognitive state, intent, and response to events are of increasing interest. Brain-computer interface (BCI) systems can make use of such knowledge to deliver relevant feedback to the user or to an observer, or within a human-machine system to increase safety and enhance overall performance. Building robust and useful BCI models from accumulated biological knowledge and available data is a major challenge, as are technical problems associated with incorporating multimodal physiological, behavioral, and contextual data that may in the future be increasingly ubiquitous. While performance of current BCI modeling methods is slowly increasing, current performance levels do not yet support widespread uses. Here we discuss the current neuroscientific questions and data processing challenges facing BCI designers and outline some promising current and future directions to address them.

We propose a model for the density of cross-spectral coefficients using normal variance mean mixtures. We show that this model density generalizes the corresponding marginal density of the complex Wishart distribution for the cross-spectral density. The maximum likelihood estimate of parameters in the distribution is derived, and examples are given from alpha brain wave sources in separated EEG data.

We derive an asymptotic Newton algorithm for quasi-maximum likelihood estimation of the ICA mixture model, using the ordinary gradient and Hessian. The probabilistic mixture framework yields an algorithm that can accommodate non-stationary environments and arbitrary source densities. We prove asymptotic stability when the source models match the true sources. An example application to EEC segmentation is given.

We present a novel experimental paradigm and data analysis methodology for studying brain dynamics during sustained-attention tasks. 256-channel EEG data were recorded while subjects participated in hour-long simulated driving sessions. Every few seconds, the vehicle drifted away from the center of the left lane, and subjects were instructed to steer back to the lane center. The error of each drifting event was measured by the maximum absolute distance from the vehicle's position at deviation onset. EEG data were analyzed using independent component analysis and time-frequency analysis. An independent component with equivalent dipole sources located bilaterally in lateral occipital cortex exhibited multi-scale brain dynamics. Tonic (~20s) alpha-band power increased in high-error compared to low-error epochs, while phasic (~1s) alpha power was suppressed briefly after deviation onset, then increased strongly just before response offset. Other components also exhibited distinct tonic and/or phasic activity patterns relating to deviation onsets or response onsets.

We present a new algorithm for maximum likelihood convolutive independent component analysis (ICA) in which components are unmixed using stable autoregressive filters determined implicitly by estimating a convolutive model of the mixing process. By introducing a convolutive mixing model for the components, we show how the order of the filters in the model can be correctly detected using Bayesian model selection. We demonstrate a framework for deconvolving a subspace of independent components in electroencephalography (EEG). Initial results suggest that in some cases, convolutive mixing may be a more realistic model for EEG signals than the instantaneous ICA model.

We have developed a new approach to studying human heart activity using independent component analysis. The electrocardiogram (ECG) is an important tool in diagnosis of heart disease. However, the normal 12-lead ECG can only record limited aspects of heart's electrical signals and mostly their interpretation relies on trained and experienced medical doctors. We have performed experiments in which heart signals were recorded in high spatial resolution. Independent component analysis was applied to the recorded signals to separate distinct temporal components of the recorded signals. The separated components were further analyzed by back-projecting their activities to the surface montage to examine each component's property. Experimental results show this to be a promising approach that can be extended to build more detailed heart activity simulations

The ill-posed nature of the MEG/EEG source localization problem requires the incorporation of prior assumptions when choosing an appropriate solution out of an infinite set of candidates. Bayesian methods are useful in this capacity because they allow these assumptions to be explicitly quantified. Recently, a number of empirical Bayesian approaches have been proposed that attempt a form of model selection by using the data to guide the search for an appropriate prior. While seemingly quite different in many respects, we apply a unifying framework based on automatic relevance determination (ARD) that elucidates various attributes of these methods and suggests directions for improvement. We also derive theoretical properties of this methodology related to convergence, local minima, and localization bias and explore connections with established algorithms.

The analysis of electroencephalographic (EEG) and magnetoencephalographic (MEG) recordings is important both for basic brain research and for medical diagnosis and treatment. Independent component analysis (ICA) is an effective method for removing artifacts and separating sources of the brain signals from these recordings. A similar approach is proving useful for analyzing functional magnetic resonance brain imaging (fMRI) data. In this paper, we outline the assumptions underlying ICA and demonstrate its application to a variety of electrical and hemodynamic recordings from the human brain.