EEG dynamic source imaging using a regularized optimization with spatio-temporal constraints

One of the most important needs in neuroimaging is brain dynamic source imaging with high spatial and temporal resolution. EEG source imaging estimates the underlying sources from EEG recordings, which provides enhanced spatial resolution with intrinsically high temporal resolution. To ensure identifiability in the underdetermined source reconstruction problem, constraints on EEG sources are essential. This paper introduces a novel method for estimating source activities based on spatio-temporal constraints and a dynamic source imaging algorithm. The method enhances time resolution by incorporating temporal evolution of neural activity into a regularization function. Additionally, two spatial regularization constraints based on L 1 and L 2 norms are applied in the transformed domain to address both focal and spread neural activities, achieved through spatial gradient and Laplacian transform. Performance evaluation, conducted quantitatively using synthetic datasets, discusses the influence of parameters such as source extent, number of sources, correlation level, and SNR level on temporal and spatial metrics. Results demonstrate that the proposed method provides superior spatial and temporal reconstructions compared to state-of-the-art inverse solutions including STRAPS, sLORETA, SBL, dSPM, and MxNE. This improvement is attributed to the simultaneous integration of transformed spatial and temporal constraints. When applied to a real auditory ERP dataset, our algorithm accurately reconstructs brain source time series and locations, effectively identifying the origins of auditory evoked potentials. In conclusion, our proposed method with spatio-temporal constraints outperforms the state-of-the-art algorithms in estimating source distribution and time courses.

Electromagnetic Source Imaging via a Data-Synthesis-Based Convolutional Encoder-Decoder Network

Electromagnetic source imaging (ESI) requires solving a highly ill-posed inverse problem. To seek a unique solution, traditional ESI methods impose various forms of priors that may not accurately reflect the actual source properties, which may hinder their broad applications. To overcome this limitation, in this article, a novel data-synthesized spatiotemporally convolutional encoder-decoder network (DST-CedNet) method is proposed for ESI. The DST-CedNet recasts ESI as a machine learning problem, where discriminative learning and latent-space representations are integrated in a CedNet to learn a robust mapping from the measured electroencephalography/magnetoencephalography (E/MEG) signals to the brain activity. In particular, by incorporating prior knowledge regarding dynamical brain activities, a novel data synthesis strategy is devised to generate large-scale samples for effectively training CedNet. This stands in contrast to traditional ESI methods where the prior information is often enforced via constraints primarily aimed for mathematical convenience. Extensive numerical experiments as well as analysis of a real MEG and epilepsy EEG dataset demonstrate that the DST-CedNet outperforms several state-of-the-art ESI methods in robustly estimating source signals under a variety of source configurations.

Electrophysiological brain imaging based on simulation-driven deep learning in the context of epilepsy

Identifying the location, the spatial extent and the electrical activity of distributed brain sources in the context of epilepsy through ElectroEncephaloGraphy (EEG) recordings is a challenging task because of the highly ill-posed nature of the underlying Electrophysiological Source Imaging (ESI) problem. To guarantee a unique solution, most existing ESI methods pay more attention to solve this inverse problem by imposing physiological constraints. This paper proposes an efficient ESI approach based on simulation-driven deep learning. Epileptic High-resolution 256-channels scalp EEG (Hr-EEG) signals are simulated in a realistic manner to train the proposed patient-specific model. More particularly, a computational neural mass model developed in our team is used to generate the temporal dynamics of the activity of each dipole while the forward problem is solved using a patient-specific three-shell realistic head model and the boundary element method. A Temporal Convolutional Network (TCN) is considered in the proposed model to capture local spatial patterns. To enable the model to observe the EEG signals from different scale levels, the multi-scale strategy is leveraged to capture the overall features and fine-grain features by adjusting the convolutional kernel size. Then, the Long Short-Term Memory (LSTM) is used to extract temporal dependencies among the computed spatial features. The performance of the proposed method is evaluated through three different scenarios of realistic synthetic interictal Hr-EEG data as well as on real interictal Hr-EEG data acquired in three patients with drug-resistant partial epilepsy, during their presurgical evaluation. A performance comparison study is also conducted with two other deep learning-based methods and four classical ESI techniques. The proposed model achieved a Dipole Localization Error (DLE) of 1.39 and Normalized Hamming Distance (NHD) of 0.28 in the case of one patch with SNR of 10 dB. In the case of two uncorrelated patches with an SNR of 10 dB, obtained DLE and NHD were respectively 1.50 and 0.28. Even in the more challenging scenario of two correlated patches with an SNR of 10 dB, the proposed approach still achieved a DLE of 3.74 and an NHD of 0.43. The results obtained on simulated data demonstrate that the proposed method outperforms the existing methods for different signal-to-noise and source configurations. The good behavior of the proposed method is also confirmed on real interictal EEG data. The robustness with respect to noise makes it a promising and alternative tool to localize epileptic brain areas and to reconstruct their electrical activities from EEG signals.

μ-STAR: A novel framework for spatio-temporal M/EEG source imaging optimized by microstates

Source imaging of Electroencephalography (EEG) and Magnetoencephalography (MEG) provides a noninvasive way of monitoring brain activities with high spatial and temporal resolution. In order to address this highly ill-posed problem, conventional source imaging models adopted spatio-temporal constraints that assume spatial stability of the source activities, neglecting the transient characteristics of M/EEG. In this work, a novel source imaging method μ-STAR that includes a microstate analysis and a spatio-temporal Bayesian model was introduced to address this problem. Specifically, the microstate analysis was applied to achieve automatic determination of time window length with quasi-stable source activity pattern for optimal reconstruction of source dynamics. Then a user-specific spatial prior and data-driven temporal basis functions were utilized to characterize the spatio-temporal information of sources within each state. The solution of the source reconstruction was obtained through a computationally efficient algorithm based upon variational Bayesian and convex analysis. The performance of the μ-STAR was first assessed through numerical simulations, where we found that the determination and inclusion of optimal temporal length in the spatio-temporal prior significantly improved the performance of source reconstruction. More importantly, the μ-STAR model achieved robust performance under various settings (i.e., source numbers/areas, SNR levels, and source depth) with fast convergence speed compared with five widely-used benchmark models (including wMNE, STV, SBL, BESTIES, & SI-STBF). Additional validations on real data were then performed on two publicly-available datasets (including block-design face-processing ERP and continuous resting-state EEG). The reconstructed source activities exhibited spatial and temporal neurophysiologically plausible results consistent with previously-revealed neural substrates, thereby further proving the feasibility of the μ-STAR model for source imaging in various applications.

A Long Short-Term Memory Network for Sparse Spatiotemporal EEG Source Imaging

EEG inverse problem is underdetermined, which poses a long standing challenge in Neuroimaging. The combination of source-imaging and analysis of cortical directional networks enables us to noninvasively explore the underlying neural processes. However, existing EEG source imaging approaches mainly focus on performing the direct inverse operation for source estimation, which will be inevitably influenced by noise and the strategy used to find the inverse solution. Here, we develop a new source imaging technique, Deep Brain Neural Network (DeepBraiNNet), for robust sparse spatiotemporal EEG source estimation. In DeepBraiNNet, considering that Recurrent Neural Network (RNN) are usually "deep" in temporal dimension and thus suitable for time sequence modelling, the RNN with Long Short-Term Memory (LSTM) is utilized to approximate the inverse operation for the lead field matrix instead of performing the direct inverse operation, which avoids the possible effect of the direct inverse operation on the underdetermined lead field matrix prone to be influenced by noise. Simulations on various source patterns and noise conditions confirmed that the proposed approach could actually recover the spatiotemporal sources well, outperforming existing state of-the-art methods. DeepBraiNNet also estimated sparse MI related activation patterns when it was applied to a real Motor Imagery dataset, consistent with other findings based on EEG and fMRI. Based on the spatiotemporal sources estimated from DeepBraiNNet, we constructed MI related cortical neural networks, which clearly exhibited strong contralateral network patterns for the two MI tasks. Consequently, DeepBraiNNet may provide an alternative way different from the conventional approaches for spatiotemporal EEG source imaging.

Sparse EEG Source Localization Using LAPPS: Least Absolute l-P (0 < p < 1) Penalized Solution

OBJECTIVE

The electroencephalographic (EEG) inverse problem is ill-posed owing to the electromagnetism Helmholtz theorem and since there are fewer observations than the unknown variables. Apart from the strong background activities (ongoing EEG), evoked EEG is also inevitably contaminated by strong outliers caused by head movements or ocular movements during recordings.

METHODS

Considering the sparse activations during high cognitive processing, we propose a novel robust EEG source imaging algorithm, LAPPS (Least Absolute -P (0 < p < 1) Penalized Solution), which employs the -loss for the residual error to alleviate the effect of outliers and another -penalty norm (p=0.5) to obtain sparse sources while suppressing Gaussian noise in EEG recordings. The resulting optimization problem is solved using a modified ADMM algorithm.

RESULTS

Simulation study was performed to recover sparse signals of randomly selected sources using LAPPS and various methods commonly used for EEG source imaging including WMNE, -norm, sLORETA and FOCUSS solution. The simulation comparison quantitatively demonstrates that LAPPS obtained the best performances in all the conducted simulations for various dipoles configurations under various SNRs on a realistic head model. Moreover, in the localization of brain neural generators in a real visual oddball experiment, LAPPS obtained sparse activations consistent with previous findings revealed by EEG and fMRI.

CONCLUSION

This study demonstrates a potentially useful sparse method for EEG source imaging, creating a platform for investigating the brain neural generators.

SIGNIFICANCE

This method alleviates the effect of noise and recovers sparse sources while maintaining a low computational complexity due to the cheap matrix-vector multiplication.

A Novel Multilayer Correlation Maximization Model for Improving CCA-Based Frequency Recognition in SSVEP Brain-Computer Interface

Multiset canonical correlation analysis (MsetCCA) has been successfully applied to optimize the reference signals by extracting common features from multiple sets of electroencephalogram (EEG) for steady-state visual evoked potential (SSVEP) recognition in brain-computer interface application. To avoid extracting the possible noise components as common features, this study proposes a sophisticated extension of MsetCCA, called multilayer correlation maximization (MCM) model for further improving SSVEP recognition accuracy. MCM combines advantages of both CCA and MsetCCA by carrying out three layers of correlation maximization processes. The first layer is to extract the stimulus frequency-related information in using CCA between EEG samples and sine-cosine reference signals. The second layer is to learn reference signals by extracting the common features with MsetCCA. The third layer is to re-optimize the reference signals set in using CCA with sine-cosine reference signals again. Experimental study is implemented to validate effectiveness of the proposed MCM model in comparison with the standard CCA and MsetCCA algorithms. Superior performance of MCM demonstrates its promising potential for the development of an improved SSVEP-based brain-computer interface.

Mixture of autoregressive modeling orders and its implication on single trial EEG classification

Autoregressive (AR) models are of commonly utilized feature types in Electroencephalogram (EEG) studies due to offering better resolution, smoother spectra and being applicable to short segments of data. Identifying correct AR's modeling order is an open challenge. Lower model orders poorly represent the signal while higher orders increase noise. Conventional methods for estimating modeling order includes Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC) and Final Prediction Error (FPE). This article assesses the hypothesis that appropriate mixture of multiple AR orders is likely to better represent the true signal compared to any single order. Better spectral representation of underlying EEG patterns can increase utility of AR features in Brain Computer Interface (BCI) systems by increasing timely & correctly responsiveness of such systems to operator's thoughts. Two mechanisms of Evolutionary-based fusion and Ensemble-based mixture are utilized for identifying such appropriate mixture of modeling orders. The classification performance of the resultant AR-mixtures are assessed against several conventional methods utilized by the community including 1) A well-known set of commonly used orders suggested by the literature, 2) conventional order estimation approaches (e.g., AIC, BIC and FPE), 3) blind mixture of AR features originated from a range of well-known orders. Five datasets from BCI competition III that contain 2, 3 and 4 motor imagery tasks are considered for the assessment. The results indicate superiority of Ensemble-based modeling order mixture and evolutionary-based order fusion methods within all datasets.

Reconstruction of Neural Activity from EEG Data Using Dynamic Spatiotemporal Constraints

We present a novel iterative regularized algorithm (IRA) for neural activity reconstruction that explicitly includes spatiotemporal constraints, performing a trade-off between space and time resolutions. For improving the spatial accuracy provided by electroencephalography (EEG) signals, we explore a basis set that describes the smooth, localized areas of potentially active brain regions. In turn, we enhance the time resolution by adding the Markovian assumption for brain activity estimation at each time period. Moreover, to deal with applications that have either distributed or localized neural activity, the spatiotemporal constraints are expressed through [Formula: see text] and [Formula: see text] norms, respectively. For the purpose of validation, we estimate the neural reconstruction performance in time and space separately. Experimental testing is carried out on artificial data, simulating stationary and non-stationary EEG signals. Also, validation is accomplished on two real-world databases, one holding Evoked Potentials and another with EEG data of focal epilepsy. Moreover, responses of functional magnetic resonance imaging for the former EEG data have been measured in advance, allowing to contrast our findings. Obtained results show that the [Formula: see text]-based IRA produces a spatial resolution that is comparable to the one achieved by some widely used sparse-based estimators of brain activity. At the same time, the [Formula: see text]-based IRA outperforms other similar smooth solutions, providing a spatial resolution that is lower than the sparse [Formula: see text]-based solution. As a result, the proposed IRA is a promising method for improving the accuracy of brain activity reconstruction.

Bayesian electromagnetic spatio-temporal imaging of extended sources with Markov Random Field and temporal basis expansion

Estimating the locations and spatial extents of brain sources poses a long-standing challenge for electroencephalography and magnetoencephalography (E/MEG) source imaging. In the present work, a novel source imaging method, Bayesian Electromagnetic Spatio-Temporal Imaging of Extended Sources (BESTIES), which is built upon a Bayesian framework that determines the spatio-temporal smoothness of source activities in a fully data-driven fashion, is proposed to address this challenge. In particular, a Markov Random Field (MRF), which can precisely capture local cortical interactions, is employed to characterize the spatial smoothness of source activities, the temporal dynamics of which are modeled by a set of temporal basis functions (TBFs). Crucially, all of the unknowns in the MRF and TBF models are learned from the data. To accomplish model inference efficiently on high-resolution source spaces, a scalable algorithm is developed to approximate the posterior distribution of the source activities, which is based on the variational Bayesian inference and convex analysis. The performance of BESTIES is assessed using both simulated and actual human E/MEG data. Compared with L₂-norm constrained methods, BESTIES is superior in reconstructing extended sources with less spatial diffusion and less localization error. By virtue of the MRF, BESTIES also overcomes the drawback of over-focal estimates in sparse constrained methods.

Sparse Bayesian Learning for Obtaining Sparsity of EEG Frequency Bands Based Feature Vectors in Motor Imagery Classification

Effective common spatial pattern (CSP) feature extraction for motor imagery (MI) electroencephalogram (EEG) recordings usually depends on the filter band selection to a large extent. Subband optimization has been suggested to enhance classification accuracy of MI. Accordingly, this study introduces a new method that implements sparse Bayesian learning of frequency bands (named SBLFB) from EEG for MI classification. CSP features are extracted on a set of signals that are generated by a filter bank with multiple overlapping subbands from raw EEG data. Sparse Bayesian learning is then exploited to implement selection of significant features with a linear discriminant criterion for classification. The effectiveness of SBLFB is demonstrated on the BCI Competition IV IIb dataset, in comparison with several other competing methods. Experimental results indicate that the SBLFB method is promising for development of an effective classifier to improve MI classification.

Three-Dimensional Innervation Zone Imaging from Multi-Channel Surface EMG Recordings

There is an unmet need to accurately identify the locations of innervation zones (IZs) of spastic muscles, so as to guide botulinum toxin (BTX) injections for the best clinical outcome. A novel 3D IZ imaging (3DIZI) approach was developed by combining the bioelectrical source imaging and surface electromyogram (EMG) decomposition methods to image the 3D distribution of IZs in the target muscles. Surface IZ locations of motor units (MUs), identified from the bipolar map of their MU action potentials (MUAPs) were employed as a prior knowledge in the 3DIZI approach to improve its imaging accuracy. The performance of the 3DIZI approach was first optimized and evaluated via a series of designed computer simulations, and then validated with the intramuscular EMG data, together with simultaneously recorded 128-channel surface EMG data from the biceps of two subjects. Both simulation and experimental validation results demonstrate the high performance of the 3DIZI approach in accurately reconstructing the distributions of IZs and the dynamic propagation of internal muscle activities in the biceps from high-density surface EMG recordings.