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

Structured noise champagne: an empirical Bayesian algorithm for electromagnetic brain imaging with structured noise

Introduction

Electromagnetic brain imaging is the reconstruction of brain activity from non-invasive recordings of electroencephalography (EEG), magnetoencephalography (MEG), and also from invasive ones such as the intracranial recording of electrocorticography (ECoG), intracranial electroencephalography (iEEG), and stereo electroencephalography EEG (sEEG). These modalities are widely used techniques to study the function of the human brain. Efficient reconstruction of electrophysiological activity of neurons in the brain from EEG/MEG measurements is important for neuroscience research and clinical applications. An enduring challenge in this field is the accurate inference of brain signals of interest while accounting for all sources of noise that contribute to the sensor measurements. The statistical characteristic of the noise plays a crucial role in the success of the brain source recovery process, which can be formulated as a sparse regression problem.

Method

In this study, we assume that the dominant environment and biological sources of noise that have high spatial correlations in the sensors can be expressed as a structured noise model based on the variational Bayesian factor analysis. To the best of our knowledge, no existing algorithm has addressed the brain source estimation problem with such structured noise. We propose to apply a robust empirical Bayesian framework for iteratively estimating the brain source activity and the statistics of the structured noise. In particular, we perform inference of the variational Bayesian factor analysis (VBFA) noise model iteratively in conjunction with source reconstruction.

Results

To demonstrate the effectiveness of the proposed algorithm, we perform experiments on both simulated and real datasets. Our algorithm achieves superior performance as compared to several existing benchmark algorithms.

Discussion

A key aspect of our algorithm is that we do not require any additional baseline measurements to estimate the noise covariance from the sensor data under scenarios such as resting state analysis, and other use cases wherein a noise or artifactual source occurs only in the active period but not in the baseline period (e.g., neuro-modulatory stimulation artifacts and speech movements).

Debiased Estimation and Inference for Spatial-Temporal EEG/MEG Source Imaging

The development of accurate electroencephalography (EEG) and magnetoencephalography (MEG) source imaging algorithm is of great importance for functional brain research and non-invasive presurgical evaluation of epilepsy. In practice, the challenge arises from the fact that the number of measurement channels is far less than the number of candidate source locations, rendering the inverse problem ill-posed. A widely used approach is to introduce a regularization term into the objective function, which inevitably biased the estimated amplitudes towards zero, leading to an inaccurate estimation of the estimator's variance. This study proposes a novel debiased EEG/MEG source imaging (DeESI) algorithm for detecting sparse brain activities, which corrects the estimation bias in signal amplitude, dipole orientation and depth. The DeESI extends the idea of group Lasso by incorporating both the matrix Frobenius norm and the L1-norm, which guarantees the estimators are only sparse over sources while maintains smoothness in time and orientation. We also derived variance of the debiased estimators for standardization and hypothesis testing. A fast alternating direction method of multipliers (ADMM) algorithm is proposed for solving the matrix form optimization problem directly without the need for vectorization. The proposed algorithm is compared with eleven existing ESI methods using simulations and an open source EEG dataset whose stimulation locations are known precisely. The DeESI exhibits the best performance in peak localization and amplitude reconstruction.

Source Imaging Method Based on Spatial Smoothing and Edge Sparsity (SISSES) and Its Application to OPM-MEG

Source estimation in magnetoencephalography (MEG) involves solving a highly ill-posed problem without a unique solution. Accurate estimation of the time course and spatial extent of the source is important for studying the mechanisms of brain activity and preoperative functional localization. Traditional methods tend to yield small-amplitude diffuse or large-amplitude focused source estimates. Recently, the structured sparsity-based source imaging algorithm has emerged as one of the most promising algorithms for improving source extent estimation. However, it suffers from a notable amplitude bias. To improve the spatiotemporal resolution of reconstructed sources, we propose a novel method called the source imaging method based on spatial smoothing and edge sparsity (SISSES). In this method, the temporal dynamics of sources are modeled using a set of temporal basis functions, and the spatial characteristics of the source are represented by a first-order Markov random field (MRF) model. In particular, sparse constraints are imposed on the MRF model residuals in the original and variation domains. Numerical simulations were conducted to validate the SISSES. The results demonstrate that SISSES outperforms benchmark methods for estimating the time course, location, and extent of patch sources. Additionally, auditory and median nerve stimulation experiments were performed using a 31-channel optically pumped magnetometer MEG system, and the SISSES was applied to the source imaging of these data. The results demonstrate that SISSES correctly identified the source regions in which brain responses occurred at different times, demonstrating its feasibility for various practical applications.

An Organ-Aware Diagnosis Framework for Radiology Report Generation

Radiology report generation (RRG) is crucial to save the valuable time of radiologists in drafting the report, therefore increasing their work efficiency. Compared to typical methods that directly transfer image captioning technologies to RRG, our approach incorporates organ-wise priors into the report generation. Specifically, in this paper, we propose Organ-aware Diagnosis (OaD) to generate diagnostic reports containing descriptions of each physiological organ. During training, we first develop a task distillation (TD) module to extract organ-level descriptions from reports. We then introduce an organ-aware report generation module that, for one thing, provides a specific description for each organ, and for another, simulates clinical situations to provide short descriptions for normal cases. Furthermore, we design an auto-balance mask loss to ensure balanced training for normal/abnormal descriptions and various organs simultaneously. Being intuitively reasonable and practically simple, our OaD outperforms SOTA alternatives by large margins on commonly used IU-Xray and MIMIC-CXR datasets, as evidenced by a 3.4% BLEU-1 improvement on MIMIC-CXR and 2.0% BLEU-2 improvement on IU-Xray.

Multi-Modal Electrophysiological Source Imaging With Attention Neural Networks Based on Deep Fusion of EEG and MEG

The process of reconstructing underlying cortical and subcortical electrical activities from Electroencephalography (EEG) or Magnetoencephalography (MEG) recordings is called Electrophysiological Source Imaging (ESI). Given the complementarity between EEG and MEG in measuring radial and tangential cortical sources, combined EEG/MEG is considered beneficial in improving the reconstruction performance of ESI algorithms. Traditional algorithms mainly emphasize incorporating predesigned neurophysiological priors to solve the ESI problem. Deep learning frameworks aim to directly learn the mapping from scalp EEG/MEG measurements to the underlying brain source activities in a data-driven manner, demonstrating superior performance compared to traditional methods. However, most of the existing deep learning approaches for the ESI problem are performed on a single modality of EEG or MEG, meaning the complementarity of these two modalities has not been fully utilized. How to fuse the EEG and MEG in a more principled manner under the deep learning paradigm remains a challenging question. This study develops a Multi-Modal Deep Fusion (MMDF) framework using Attention Neural Networks (ANN) to fully leverage the complementary information between EEG and MEG for solving the ESI inverse problem, which is termed as MMDF-ANN. Specifically, our proposed brain source imaging approach consists of four phases, including feature extraction, weight generation, deep feature fusion, and source mapping. Our experimental results on both synthetic dataset and real dataset demonstrated that using a fusion of EEG and MEG can significantly improve the source localization accuracy compared to using a single-modality of EEG or MEG. Compared to the benchmark algorithms, MMDF-ANN demonstrated good stability when reconstructing sources with extended activation areas and situations of EEG/MEG measurements with a low signal-to-noise ratio.

VSSI-GGD: A Variation Sparse EEG Source Imaging Approach Based on Generalized Gaussian Distribution

Electroencephalographic (EEG) source imaging (ESI) is a powerful method for studying brain functions and surgical resection of epileptic foci. However, accurately estimating the location and extent of brain sources remains challenging due to noise and background interference in EEG signals. To reconstruct extended brain sources, we propose a new ESI method called Variation Sparse Source Imaging based on Generalized Gaussian Distribution (VSSI-GGD). VSSI-GGD uses the generalized Gaussian prior as a sparse constraint on the spatial variation domain and embeds it into the Bayesian framework for source estimation. Using a variational technique, we approximate the intractable true posterior with a Gaussian density. Through convex analysis, the Bayesian inference problem is transformed entirely into a series of regularized L2p -norm ( ) optimization problems, which are efficiently solved with the ADMM algorithm. Imaging results of numerical simulations and human experimental dataset analysis reveal the superior performance of VSSI-GGD, which provides higher spatial resolution with clear boundaries compared to benchmark algorithms. VSSI-GGD can potentially serve as an effective and robust spatiotemporal EEG source imaging method. The source code of VSSI-GGD is available at https://github.com/Mashirops/VSSI-GGD.git.

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.

Deep learning based source imaging provides strong sublobar localization of epileptogenic zone from MEG interictal spikes

Electromagnetic source imaging (ESI) offers unique capability of imaging brain dynamics for studying brain functions and aiding the clinical management of brain disorders. Challenges exist in ESI due to the ill-posedness of the inverse problem and thus the need of modeling the underlying brain dynamics for regularizations. Advances in generative models provide opportunities for more accurate and realistic source modeling that could offer an alternative approach to ESI for modeling the underlying brain dynamics beyond equivalent physical source models. However, it is not straightforward to explicitly formulate the knowledge arising from these generative models within the conventional ESI framework. Here we investigate a novel source imaging framework based on mesoscale neuronal modeling and deep learning (DL) that can learn the sensor-source mapping relationship directly from MEG data for ESI. Two DL-based ESI models were trained based on data generated by neural mass models and either generic or personalized head models. The robustness of the two DL models was evaluated by systematic computer simulations and clinical validation in a cohort of 29 drug-resistant focal epilepsy patients who underwent intracranial EEG (iEEG) evaluation or surgical resection. Results estimated from pre-operative MEG interictal spikes were quantified using the overlap with resection regions and the distance to the seizure-onset zone (SOZ) defined by iEEG recordings. The DL-based ESI provided robust results when no personalized head geometry is considered, reaching a spatial dispersion of 21.90 ± 19.03 mm, sublobar concordance of 83 %, and sublobar sensitivity and specificity of 66 and 97 % respectively. When using personalized head geometry derived from individual patients' MRI in the training data, personalized DL-based ESI model can further improve the performance and reached a spatial dispersion of 8.19 ± 8.14 mm, sublobar concordance of 93 %, and sublobar sensitivity and specificity of 77 and 99 % respectively. When compared to the SOZ, the localization error of the personalized approach is 15.78 ± 5.54 mm, outperforming the conventional benchmarks. This work demonstrates that combining generative models and deep learning enables an accurate and robust imaging of epileptogenic zone from MEG recordings with strong sublobar precision, suggesting its added value to enhancing MEG source localization and imaging, and to epilepsy source localization and other clinical applications.

Patient-specific solution of the electrocorticography forward problem in deforming brain

Invasive intracranial electroencephalography (iEEG), or electrocorticography (ECoG), measures electric potential directly on the surface of the brain and can be used to inform treatment planning for epilepsy surgery. Combined with numerical modeling it can further improve accuracy of epilepsy surgery planning. Accurate solution of the iEEG forward problem, which is a crucial prerequisite for solving the iEEG inverse problemin epilepsy seizure onset zone localization, requires accurate representation of the patient's brain geometry and tissue electrical conductivity after implantation of electrodes. However, implantation of subdural grid electrodes causes the brain to deform, which invalidates preoperatively acquired image data. Moreover, postoperative magnetic resonance imaging (MRI) is incompatible with implanted electrodes and computed tomography (CT) has insufficient range of soft tissue contrast, which precludes both MRI and CT from being used to obtain the deformed postoperative geometry. In this paper, we present a biomechanics-based image warping procedure using preoperative MRI for tissue classification and postoperative CT for locating implanted electrodes to perform non-rigid registration of the preoperative image data to the postoperative configuration. We solve the iEEG forward problem on the predicted postoperative geometry using the finite element method (FEM) which accounts for patient-specific inhomogeneity and anisotropy of tissue conductivity. Results for the simulation of a current source in the brain show large differences in electric potential predicted by the models based on the original images and the deformed images corresponding to the brain geometry deformed by placement of invasive electrodes. Computation of the lead field matrix (useful for solution of the iEEG inverse problem) also showed significant differences between the different models. The results suggest that rapid and accurate solution of the forward problem in a deformed brain for a given patient is achievable.

EEG Source Imaging using GANs with Deep Image Prior

Brain source localization from electroencephalogram (EEG) signals is an challenging problem for noninvasively localizing the brain activity. Conventional methods use handcrafted regularization terms based on neural-physiological assumptions by exploiting the spatial-temporal structure on the source signals. In recent years, deep learning frameworks have demonstrated superior performance for solving the inverse problems in the natural and medical imaging field. This study proposes a novel unsupervised learning training-free framework based on Generative Adversarial Networks and deep image prior (GANs-DIP) as a generative model simulating spatially structured source signal. The proposed framework can faithfully recover extended source patches activation patterns of the brain in an unsupervised manner. Numerical experiments on a realistic brain model are performed under different levels of signal-to-noise ratio (SNR). The proposed model shows satisfactory performance in recovering the underlying source activation.

Biophysical Model: A Promising Method in the Study of the Mechanism of Propofol: A Narrative Review

The physiological and neuroregulatory mechanism of propofol is largely based on very limited knowledge. It is one of the important puzzling issues in anesthesiology and is of great value in both scientific and clinical fields. It is acknowledged that neural networks which are comprised of a number of neural circuits might be involved in the anesthetic mechanism. However, the mechanism of this hypothesis needs to be further elucidated. With the progress of artificial intelligence, it is more likely to solve this problem through using artificial neural networks to perform temporal waveform data analysis and to construct biophysical computational models. This review focuses on current knowledge regarding the anesthetic mechanism of propofol, an intravenous general anesthetic, by constructing biophysical computational models.

A Graph Fourier Transform Based Bidirectional Long Short-Term Memory Neural Network for Electrophysiological Source Imaging

Electrophysiological source imaging (ESI) refers to the process of reconstructing underlying activated sources on the cortex given the brain signal measured by Electroencephalography (EEG) or Magnetoencephalography (MEG). Due to the ill-posed nature of ESI, solving ESI requires the design of neurophysiologically plausible regularization or priors to guarantee a unique solution. Recovering focally extended sources is more challenging, and traditionally uses a total variation regularization to promote spatial continuity of the activated sources. In this paper, we propose to use graph Fourier transform (GFT) based bidirectional long-short term memory (BiLSTM) neural network to solve the ESI problem. The GFT delineates the 3D source space into spatially high, medium and low frequency subspaces spanned by corresponding eigenvectors. The low frequency components can naturally serve as a spatially low-band pass filter to reconstruct extended areas of source activation. The BiLSTM is adopted to learn the mapping relationship between the projection of low-frequency graph space and the recorded EEG. Numerical results show the proposed GFT-BiLSTM outperforms other benchmark algorithms in synthetic data under varied signal-to-noise ratios (SNRs). Real data experiments also demonstrate its capability of localizing the epileptogenic zone of epilepsy patients with good accuracy.

Imaging somatosensory cortex responses measured by OPM-MEG: Variational free energy-based spatial smoothing estimation approach

In recent years, optically pumped magnetometer (OPM)-based magnetoencephalography (MEG) has shown potential for analyzing brain activity. It has a flexible sensor configuration and comparable sensitivity to conventional SQUID-MEG. We constructed a 32-channel OPM-MEG system and used it to measure cortical responses to median and ulnar nerve stimulations. Traditional magnetic source imaging methods tend to blur the spatial extent of sources. Accurate estimation of the spatial size of the source is important for studying the organization of brain somatotopy and for pre-surgical functional mapping. We proposed a new method called variational free energy-based spatial smoothing estimation (FESSE) to enhance the accuracy of mapping somatosensory cortex responses. A series of computer simulations based on the OPM-MEG showed better performance than the three types of competing methods under different levels of signal-to-noise ratios, source patch sizes, and co-registration errors. FESSE was then applied to the source imaging of the OPM-MEG experimental data.