Sparse EEG Source Localization Using Bernoulli Laplacian Priors

High-resolution EEG source localization in personalized segmentation-free head model with multi-dipole fitting

Objective. Electroencephalograms (EEGs) are often used to monitor brain activity. Several source localization methods have been proposed to estimate the location of brain activity corresponding to EEG readings. However, only a few studies evaluated source localization accuracy from measured EEG using personalized head models in a millimeter resolution. In this study, based on a volume conductor analysis of a high-resolution personalized human head model constructed from magnetic resonance images, a finite difference method was used to solve the forward problem and to reconstruct the field distribution.Approach. We used a personalized segmentation-free head model developed using machine learning techniques, in which the abrupt change of electrical conductivity occurred at the tissue interface is suppressed. Using this model, a smooth field distribution was obtained to address the forward problem. Next, multi-dipole fitting was conducted using EEG measurements for each subject (N= 10 male subjects, age: 22.5 ± 0.5), and the source location and electric field distribution were estimated.Main results.For measured somatosensory evoked potential for electrostimulation to the wrist, a multi-dipole model with lead field matrix computed with the volume conductor model was found to be superior than a single dipole model when using personalized segmentation-free models (6/10). The correlation coefficient between measured and estimated scalp potentials was 0.89 for segmentation-free head models and 0.71 for conventional segmented models. The proposed method is straightforward model development and comparable localization difference of the maximum electric field from the target wrist reported using fMR (i.e. 16.4 ± 5.2 mm) in previous study. For comparison, DUNEuro based on sLORETA was (EEG: 17.0 ± 4.0 mm). In addition, somatosensory evoked magnetic fields obtained by Magnetoencephalography was 25.3 ± 8.5 mm using three-layer sphere and sLORETA.Significance. For measured EEG signals, our procedures using personalized head models demonstrated that effective localization of the somatosensory cortex, which is located in a non-shallower cortex region. This method may be potentially applied for imaging brain activity located in other non-shallow regions.

Multiple Dipole Source Position and Orientation Estimation Using Non-Invasive EEG-like Signals

The problem of precisely estimating the position and orientation of multiple dipoles using synthetic EEG signals is considered in this paper. After determining a proper forward model, a nonlinear constrained optimization problem with regularization is solved, and the results are compared with a widely used research code, namely EEGLAB. A thorough sensitivity analysis of the estimation algorithm to the parameters (such as the number of samples and sensors) in the assumed signal measurement model is conducted. To confirm the efficacy of the proposed source identification algorithm on any category of data sets, three different kinds of data-synthetic model data, visually evoked clinical EEG data, and seizure clinical EEG data are used. Furthermore, the algorithm is tested on both the spherical head model and the realistic head model based on the MNI coordinates. The numerical results and comparisons with the EEGLAB show very good agreement, with little pre-processing required for the acquired data.

Accurate emotion recognition using Bayesian model based EEG sources as dynamic graph convolutional neural network nodes

Due to the effect of emotions on interactions, interpretations, and decisions, automatic detection and analysis of human emotions based on EEG signals has an important role in the treatment of psychiatric diseases. However, the low spatial resolution of EEG recorders poses a challenge. In order to overcome this problem, in this paper we model each emotion by mapping from scalp sensors to brain sources using Bernoulli–Laplace-based Bayesian model. The standard low-resolution electromagnetic tomography (sLORETA) method is used to initialize the source signals in this algorithm. Finally, a dynamic graph convolutional neural network (DGCNN) is used to classify emotional EEG in which the sources of the proposed localization model are considered as the underlying graph nodes. In the proposed method, the relationships between the EEG source signals are encoded in the DGCNN adjacency matrix. Experiments on our EEG dataset recorded at the Brain-Computer Interface Research Laboratory, University of Tabriz as well as publicly available SEED and DEAP datasets show that brain source modeling by the proposed algorithm significantly improves the accuracy of emotion recognition, such that it achieve a classification accuracy of 99.25% during the classification of the two classes of positive and negative emotions. These results represent an absolute 1–2% improvement in terms of classification accuracy over subject-dependent and subject-independent scenarios over the existing approaches.

High-Resolution EEG Source Localization in Segmentation-Free Head Models Based on Finite-Difference Method and Matching Pursuit Algorithm

Electroencephalogram (EEG) is a method to monitor electrophysiological activity on the scalp, which represents the macroscopic activity of the brain. However, it is challenging to identify EEG source regions inside the brain based on data measured by a scalp-attached network of electrodes. The accuracy of EEG source localization significantly depends on the type of head modeling and inverse problem solver. In this study, we adopted different models with a resolution of 0.5 mm to account for thin tissues/fluids, such as the cerebrospinal fluid (CSF) and dura. In particular, a spatially dependent conductivity (segmentation-free) model created using deep learning was developed and used for more realist representation of electrical conductivity. We then adopted a multi-grid-based finite-difference method (FDM) for forward problem analysis and a sparse-based algorithm to solve the inverse problem. This enabled us to perform efficient source localization using high-resolution model with a reasonable computational cost. Results indicated that the abrupt spatial change in conductivity, inherent in conventional segmentation-based head models, may trigger source localization error accumulation. The accurate modeling of the CSF, whose conductivity is the highest in the head, was an important factor affecting localization accuracy. Moreover, computational experiments with different noise levels and electrode setups demonstrate the robustness of the proposed method with segmentation-free head model.

Probabilistic Structure Learning for EEG/MEG Source Imaging With Hierarchical Graph Priors

Brain source imaging is an important method for noninvasively characterizing brain activity using Electroencephalogram (EEG) or Magnetoencephalography (MEG) recordings. Traditional EEG/MEG Source Imaging (ESI) methods usually assume the source activities at different time points are unrelated, and do not utilize the temporal structure in the source activation, making the ESI analysis sensitive to noise. Some methods may encourage very similar activation patterns across the entire time course and may be incapable of accounting the variation along the time course. To effectively deal with noise while maintaining flexibility and continuity among brain activation patterns, we propose a novel probabilistic ESI model based on a hierarchical graph prior. Under our method, a spanning tree constraint ensures that activity patterns have spatiotemporal continuity. An efficient algorithm based on an alternating convex search is presented to solve the resulting problem of the proposed model with guaranteed convergence. Comprehensive numerical studies using synthetic data on a realistic brain model are conducted under different levels of signal-to-noise ratio (SNR) from both sensor and source spaces. We also examine the EEG/MEG datasets in two real applications, in which our ESI reconstructions are neurologically plausible. All the results demonstrate significant improvements of the proposed method over benchmark methods in terms of source localization performance, especially at high noise levels.

A Novel Bayesian Approach for EEG Source Localization

We propose a new method for EEG source localization. An efficient solution to this problem requires choosing an appropriate regularization term in order to constraint the original problem. In our work, we adopt the Bayesian framework to place constraints; hence, the regularization term is closely connected to the prior distribution. More specifically, we propose a new sparse prior for the localization of EEG sources. The proposed prior distribution has sparse properties favoring focal EEG sources. In order to obtain an efficient algorithm, we use the variational Bayesian (VB) framework which provides us with a tractable iterative algorithm of closed-form equations. Additionally, we provide extensions of our method in cases where we observe group structures and spatially extended EEG sources. We have performed experiments using synthetic EEG data and real EEG data from three publicly available datasets. The real EEG data are produced due to the presentation of auditory and visual stimulus. We compare the proposed method with well-known approaches of EEG source localization and the results have shown that our method presents state-of-the-art performance, especially in cases where we expect few activated brain regions. The proposed method can effectively detect EEG sources in various circumstances. Overall, the proposed sparse prior for EEG source localization results in more accurate localization of EEG sources than state-of-the-art approaches.

Attentional Processes in Children With Attentional Problems or Reading Difficulties as Revealed Using Brain Event-Related Potentials and Their Source Localization

Visual attention-related processes include three functional sub-processes: alerting, orienting, and inhibition. We examined these sub-processes using reaction times, event-related potentials (ERPs), and their neuronal source activations during the Attention Network Test (ANT) in control children, attentional problems (AP) children, and reading difficulties (RD) children. During the ANT, electroencephalography was measured using 128 electrodes on three groups of Finnish sixth-graders aged 12–13 years (control = 77; AP = 15; RD = 23). Participants were asked to detect the direction of a middle target fish within a group of five fish. The target stimulus was either preceded by a cue (center, double, or spatial), or without a cue, to manipulate the alerting and orienting sub-processes of attention. The direction of the target fish was either congruent or incongruent in relation to the flanker fish, thereby manipulating the inhibition sub-processes of attention. Reaction time performance showed no differences between groups in alerting, orienting, and inhibition effects. The group differences in ERPs were only found at the source level. Neuronal source analysis in the AP children revealed a larger alerting effect (double-cued vs. non-cued target stimuli) than control and RD children in the left occipital lobe. Control children showed a smaller orienting effect (spatially cued vs. center-cued target stimuli) in the left occipital lobe than AP and RD children. No group differences were found for the neuronal sources related to the inhibition effect. The neuronal activity differences related to sub-processes of attention in the AP and RD groups suggest different underlying mechanisms for attentional and reading problems.

Attentional processes in typically developing children as revealed using brain event-related potentials and their source localization in Attention Network Test

Attention-related processes include three functional sub-components: alerting, orienting, and inhibition. We investigated these components using EEG-based, brain event-related potentials and their neuronal source activations during the Attention Network Test in typically developing school-aged children. Participants were asked to detect the swimming direction of the centre fish in a group of five fish. The target stimulus was either preceded by a cue (centre, double, or spatial) or no cue. An EEG using 128 electrodes was recorded for 83 children aged 12-13 years. RTs showed significant effects across all three sub-components of attention. Alerting and orienting (responses to double vs non-cued target stimulus and spatially vs centre-cued target stimulus, respectively) resulted in larger N1 amplitude, whereas inhibition (responses to incongruent vs congruent target stimulus) resulted in larger P3 amplitude. Neuronal source activation for the alerting effect was localized in the right anterior temporal and bilateral occipital lobes, for the orienting effect bilaterally in the occipital lobe, and for the inhibition effect in the medial prefrontal cortex and left anterior temporal lobe. Neuronal sources of ERPs revealed that sub-processes related to the attention network are different in children as compared to earlier adult fMRI studies, which was not evident from scalp ERPs.

Certainty-Based Reduced Sparse Solution for Dense Array EEG Source Localization

The EEG source localization is an ill-posed problem. It involves estimation of the sources which outnumbers the number of measurements. For a given measurement at a given time all sources are not active this makes the problem as sparse inversion problem. This paper presents a new approach for dense array EEG source localization. This paper aims at reducing the solution space to only most certain sources and thereby reducing the problem of ill-posedness. This employs a two-stage method, where the first stage finds the most certain sources that are likely to produce the observed EEG by using a statistical measure of sources, the second stage solves the inverse problem by restricting the solution space to only most certain sources and their neighbors. This reduces the solution space for other source localization methods hence improvise their accuracy in localizing the active neurological sources in the brain. This method has been validated and applied to real 256 channel data and the results were analyzed.

Subject-based discriminative sparse representation model for detection of concealed information

BACKGROUND AND OBJECTIVES

The use of machine learning approaches in concealed information test (CIT) plays a key role in the progress of this neurophysiological field. In this paper, we presented a new machine learning method for CIT in which each subject is considered independent of the others. The main goal of this study is to adapt the discriminative sparse models to be applicable for subject-based concealed information test.

METHODS

In order to provide sufficient discriminability between guilty and innocent subjects, we introduced a novel discriminative sparse representation model and its appropriate learning methods. For evaluation of the method forty-four subjects participated in a mock crime scenario and their EEG data were recorded. As the model input, in this study the recurrence plot features were extracted from single trial data of different stimuli. Then the extracted feature vectors were reduced using statistical dependency method. The reduced feature vector went through the proposed subject-based sparse model in which the discrimination power of sparse code and reconstruction error were applied simultaneously.

RESULTS

Experimental results showed that the proposed approach achieved better performance than other competing discriminative sparse models. The classification accuracy, sensitivity and specificity of the presented sparsity-based method were about 93%, 91% and 95% respectively.

CONCLUSIONS

Using the EEG data of a single subject in response to different stimuli types and with the aid of the proposed discriminative sparse representation model, one can distinguish guilty subjects from innocent ones. Indeed, this property eliminates the necessity of several subject EEG data in model learning and decision making for a specific subject.

Extraction of features from sleep EEG for Bayesian assessment of brain development

Brain development can be evaluated by experts analysing age-related patterns in sleep electroencephalograms (EEG). Natural variations in the patterns, noise, and artefacts affect the evaluation accuracy as well as experts’ agreement. The knowledge of predictive posterior distribution allows experts to estimate confidence intervals within which decisions are distributed. Bayesian approach to probabilistic inference has provided accurate estimates of intervals of interest. In this paper we propose a new feature extraction technique for Bayesian assessment and estimation of predictive distribution in a case of newborn brain development assessment. The new EEG features are verified within the Bayesian framework on a large EEG data set including 1,100 recordings made from newborns in 10 age groups. The proposed features are highly correlated with brain maturation and their use increases the assessment accuracy.

Bayesian EEG source localization using a structured sparsity prior

This paper deals with EEG source localization. The aim is to perform spatially coherent focal localization and recover temporal EEG waveforms, which can be useful in certain clinical applications. A new hierarchical Bayesian model is proposed with a multivariate Bernoulli Laplacian structured sparsity prior for brain activity. This distribution approximates a mixed ℓ₂₀ pseudo norm regularization in a Bayesian framework. A partially collapsed Gibbs sampler is proposed to draw samples asymptotically distributed according to the posterior of the proposed Bayesian model. The generated samples are used to estimate the brain activity and the model hyperparameters jointly in an unsupervised framework. Two different kinds of Metropolis-Hastings moves are introduced to accelerate the convergence of the Gibbs sampler. The first move is based on multiple dipole shifts within each MCMC chain, whereas the second exploits proposals associated with different MCMC chains. Experiments with focal synthetic data shows that the proposed algorithm is more robust and has a higher recovery rate than the weighted ℓ₂₁ mixed norm regularization. Using real data, the proposed algorithm finds sources that are spatially coherent with state of the art methods, namely a multiple sparse prior approach and the Champagne algorithm. In addition, the method estimates waveforms showing peaks at meaningful timestamps. This information can be valuable for activity spread characterization.