MEG language lateralization in partial epilepsy using dSPM of auditory event-related fields

Magnetoencephalography for the pediatric population, indications, acquisition and interpretation for the clinician

Magnetoencephalography (MEG) is an imaging technique that enables the assessment of cortical activity via direct measures of neurophysiology. It is a non-invasive and passive technique that is completely painless. MEG has gained increasing prominence in the field of pediatric neuroimaging. This dedicated review article for the pediatric population summarizes the fundamental technical and clinical aspects of MEG for the clinician. We discuss methods tailored for children to improve data quality, including child-friendly MEG facility environments and strategies to mitigate motion artifacts. We provide an in-depth overview on accurate localization of neural sources and different analysis methods, as well as data interpretation. The contemporary platforms and approaches of two quaternary pediatric referral centers are illustrated, shedding light on practical implementations in clinical settings. Finally, we describe the expanding clinical applications of MEG, including its pivotal role in presurgical evaluation of epilepsy patients, presurgical mapping of eloquent cortices (somatosensory and motor cortices, visual and auditory cortices, lateralization of language), its emerging relevance in autism spectrum disorder research and potential future clinical applications, and its utility in assessing mild traumatic brain injury. In conclusion, this review serves as a comprehensive resource of clinicians as well as researchers, offering insights into the evolving landscape of pediatric MEG. It discusses the importance of technical advancements, data acquisition strategies, and expanding clinical applications in harnessing the full potential of MEG to study neurological conditions in the pediatric population.

Magnetoencephalography Language Mapping Using Auditory Memory Retrieval and Silent Repeating Task

PURPOSE

The study aims to (1) examine the spatiotemporal map of magnetoencephalography-evoked responses during an Auditory Memory Retrieval and Silent Repeating (AMRSR) task, and determine the hemispheric dominance for language, and (2) evaluate the accuracy of the AMRSR task in Wernicke and Broca area localization.

METHODS

In 30 patients with brain tumors and/or epilepsies, the AMRSR task was used to evoke magnetoencephalography responses. We applied Fast VEctor-based Spatial-Temporal Analyses with minimum L1-norm source imaging method to the magnetoencephalography responses for localizing the brain areas evoked by the AMRSR task.

RESULTS

The Fast-VEctor-based Spatial-Temporal Analysis found consistent activation in the posterior superior temporal gyrus around 300 to 500 ms, and another activation in the frontal cortex (pars opercularis and/or pars triangularis) around 600 to 900 ms, which were localized to the Wernicke area (BA 22) and Broca area (BA 44 and BA 45), respectively. The language-dominant hemispheric laterization elicited by the AMRSR task was comparable with the result from an Auditory Dichotic task result given to the same patient, with the exception that AMRSR is more sensitive on bilateral language laterization cases on finding the Wernicke and Broca areas.

CONCLUSIONS

For all patients who successfully finished the AMRSR task, Fast-VEctor-based Spatial-Temporal Analysis could establish accurate and robust localizations of Broca and Wernicke area and determine hemispheric dominance. For subjects with normal auditory functionality, the AMRSR paradigm evaluation showed significant promise in providing reliable assessments of cerebral language dominance and language network localization.

Language MEG predicts postoperative verbal memory change in left mesial temporal lobe epilepsy

OBJECTIVE

To clarify whether preoperative language magnetoencephalography (MEG) predicts postoperative verbal memory (VM) changes in left mesial temporal lobe epilepsy (LMTLE).

METHODS

We reviewed 18 right-handed patients with LMTLE who underwent anterior temporal lobectomy or selective amygdala hippocampectomy, 12 with (HS+) and 6 without hippocampal sclerosis (HS-). Patients underwent neuropsychological assessment before and after surgery. MEG was measured with an auditory verbal learning task in patients preoperatively and in 15 right-handed controls. Dynamic statistical parametric mapping (dSPM) was used for source imaging of task-related activity. Language laterality index (LI) was calculated by z-score of dSPM in language-related regions. LI in the region of HS+ and HS- was compared to controls. The correlation between LI and postoperative VM change was assessed in HS+ and HS-.

RESULTS

Preoperative LI in supramarginal gyrus showed greater right-shifted lateralization in both HS+ and HS- than in controls. Right-shifted LI in supramarginal gyrus was correlated with postoperative VM increase in HS+ (p = 0.019), but not in HS-.

CONCLUSIONS

Right-shifted language lateralization in dSPM of MEG signals may predict favorable VM outcome in HS+ of LMTLE.

SIGNIFICANCE

Findings warrant further investigation of the relation between regional language laterality index and postoperative verbal memory changes.

MEG language mapping using a novel automatic ECD algorithm in comparison with MNE, dSPM, and DICS beamformer

Introduction

The single equivalent current dipole (sECD) is the standard clinical procedure for presurgical language mapping in epilepsy using magnetoencephalography (MEG). However, the sECD approach has not been widely used in clinical assessments, mainly because it requires subjective judgements in selecting several critical parameters. To address this limitation, we developed an automatic sECD algorithm (AsECDa) for language mapping.

Methods

The localization accuracy of the AsECDa was evaluated using synthetic MEG data. Subsequently, the reliability and efficiency of AsECDa were compared to three other common source localization methods using MEG data recorded during two sessions of a receptive language task in 21 epilepsy patients. These methods include minimum norm estimation (MNE), dynamic statistical parametric mapping (dSPM), and dynamic imaging of coherent sources (DICS) beamformer.

Results

For the synthetic single dipole MEG data with a typical signal-to-noise ratio, the average localization error of AsECDa was less than 2 mm for simulated superficial and deep dipoles. For the patient data, AsECDa showed better test-retest reliability (TRR) of the language laterality index (LI) than MNE, dSPM, and DICS beamformer. Specifically, the LI calculated with AsECDa revealed excellent TRR between the two MEG sessions across all patients (Cor = 0.80), while the LI for MNE, dSPM, DICS-event-related desynchronization (ERD) in the alpha band, and DICS-ERD in the low beta band ranged lower (Cor = 0.71, 0.64, 0.54, and 0.48, respectively). Furthermore, AsECDa identified 38% of patients with atypical language lateralization (i.e., right lateralization or bilateral), compared to 73%, 68%, 55%, and 50% identified by DICS-ERD in the low beta band, DICS-ERD in the alpha band, MNE, and dSPM, respectively. Compared to other methods, AsECDa's results were more consistent with previous studies that reported atypical language lateralization in 20-30% of epilepsy patients.

Discussion

Our study suggests that AsECDa is a promising approach for presurgical language mapping, and its fully automated nature makes it easy to implement and reliable for clinical evaluations.

Localization of beta power decrease as measure for lateralization in pre-surgical language mapping with magnetoencephalography, compared with functional magnetic resonance imaging and validated by Wada test

Objective: Atypical patterns of language lateralization due to early reorganizational processes constitute a challenge in the pre-surgical evaluation of patients with pharmaco-resistant epilepsy. There is no consensus on an optimal analysis method used for the identification of language dominance in MEG. This study examines the concordance between MEG source localization of beta power desynchronization and fMRI with regard to lateralization and localization of expressive and receptive language areas using a visual verb generation task. Methods: Twenty-five patients with pharmaco-resistant epilepsy, including six patients with atypical language lateralization, and ten right-handed controls obtained MEG and fMRI language assessment. Fourteen patients additionally underwent the Wada test. We analyzed MEG beta power desynchronization in sensor (controls) and source space (patients and controls). Beta power decrease between 13 and 35 Hz was localized applying Dynamic Imaging of Coherent Sources Beamformer technique. Statistical inferences were grounded on cluster-based permutation testing for single subjects. Results: Event-related desynchronization of beta power in MEG was seen within the language-dominant frontal and temporal lobe and within the premotor cortex. Our analysis pipeline consistently yielded left language dominance with high laterality indices in controls. Language lateralization in MEG and Wada test agreed in all 14 patients for inferior frontal, temporal and parietal language areas (Cohen's Kappa = 1, p < 0.001). fMRI agreed with Wada test in 12 out of 14 cases (85.7%) for Broca's area (Cohen's Kappa = 0.71, p = 0.024), while the agreement for temporal and temporo-parietal language areas were non-significant. Concordance between MEG and fMRI laterality indices was highest within the inferior frontal gyrus, with an agreement in 19/24 cases (79.2%), and non-significant for Wernicke's area. Spatial agreement between fMRI and MEG varied considerably between subjects and brain regions with the lowest Euclidean distances within the inferior frontal region of interest. Conclusion: Localizing the desynchronization of MEG beta power using a verb generation task is a promising tool for the identification of language dominance in the pre-surgical evaluation of epilepsy patients. The overall agreement between MEG and fMRI was lower than expected and might be attributed to differences within the baseline condition. A larger sample size and an adjustment of the experimental designs are needed to draw further conclusions.

Effects of individualized brain anatomies and EEG electrode positions on inferred activity of the primary auditory cortex

Due to its high temporal resolution and non-invasive nature, electroencephalography (EEG) is considered a method of great value for the field of auditory cognitive neuroscience. In performing source space analyses, localization accuracy poses a bottleneck, which precise forward models based on individualized attributes such as subject anatomy or electrode locations aim to overcome. Yet acquiring anatomical images or localizing EEG electrodes requires significant additional funds and processing time, making it an oftentimes inaccessible asset. Neuroscientific software offers template solutions, on which analyses can be based. For localizing the source of auditory evoked responses, we here compared the results of employing such template anatomies and electrode positions versus the subject-specific ones, as well as combinations of the two. All considered cases represented approaches commonly used in electrophysiological studies. We considered differences between two commonly used inverse solutions (dSPM, sLORETA) and targeted the primary auditory cortex; a notoriously small cortical region that is located within the lateral sulcus, thus particularly prone to errors in localization. Through systematical comparison of early evoked component metrics and spatial leakage, we assessed how the individualization steps impacted the analyses outcomes. Both electrode locations as well as subject anatomies were found to have an effect, which though varied based on the configuration considered. When comparing the inverse solutions, we moreover found that dSPM more consistently benefited from individualization of subject morphologies compared to sLORETA, suggesting it to be the better choice for auditory cortex localization.

Contributions of electrophysiology for identifying cortical language systems in patients with epilepsy

A crucial element of the surgical treatment of medically refractory epilepsy is to delineate cortical areas that must be spared in order to avoid clinically relevant neurological and neuropsychological deficits postoperatively. For each patient, this typically necessitates determining the language lateralization between hemispheres and language localization within hemisphere. Understanding cortical language systems is complicated by two primary challenges: the extent of the neural tissue involved and the substantial variability across individuals, especially in pathological populations. We review the contributions made through the study of electrophysiological activity to address these challenges. These contributions are based on the techniques of magnetoencephalography (MEG), intracerebral recordings, electrical-cortical stimulation (ECS), and the electrovideo analyses of seizures and their semiology. We highlight why no single modality alone is adequate to identify cortical language systems and suggest avenues for improving current practice.

Mapping critical hubs of receptive and expressive language using MEG: A comparison against fMRI

The complexity of the widespread language network makes it challenging for accurate localization and lateralization. Using large-scale connectivity and graph-theoretical analyses of task-based magnetoencephalography (MEG), we aimed to provide robust representations of receptive and expressive language processes, comparable with spatial profiles of corresponding functional magnetic resonance imaging (fMRI). We examined MEG and fMRI data from 12 healthy young adults (age 20-37 years) completing covert auditory word-recognition task (WRT) and covert auditory verb-generation task (VGT). For MEG language mapping, broadband (3-30 Hz) beamformer sources were estimated, voxel-level connectivity was quantified using phase locking value, and highly connected hubs were characterized using eigenvector centrality graph measure. fMRI data were analyzed using a classic general linear model approach. A laterality index (LI) was computed for 20 language-specific frontotemporal regions for both MEG and fMRI. MEG network analysis showed bilateral and symmetrically distributed hubs within the left and right superior temporal gyrus (STG) during WRT and predominant hubs in left inferior prefrontal gyrus (IFG) during VGT. MEG and fMRI localization maps showed high correlation values within frontotemporal regions during WRT and VGT (r = 0.63, 0.74, q < 0.05, respectively). Despite good concordance in localization, notable discordances were observed in lateralization between MEG and fMRI. During WRT, MEG favored a left-hemispheric dominance of left STG (LI = 0.25 ± 0.22) whereas fMRI supported a bilateral representation of STG (LI = 0.08 ± 0.2). Laterality of MEG and fMRI during VGT consistently showed a strong asymmetry in left IFG regions (MEG-LI = 0.45 ± 0.35 and fMRI-LI = 0.46 ± 0.13). Our results demonstrate the utility of a large-scale connectivity and graph theoretical analyses for robust identification of language-specific regions. MEG hubs are in great agreement with the literature in revealing with canonical and extra-canonical language sites, thus providing additional support for the underlying topological organization of receptive and expressive language cortices. Discordances in lateralization may emphasize the need for multimodal integration of MEG and fMRI to obtain an excellent predictive value in a heterogeneous healthy population and patients with neurosurgical conditions.

Clinical application of advanced MR methods in children: points to consider

The application of both functional MRI and diffusion MR tractography prior to a neurosurgical operation is well established in adults, but less so in children, for several reasons. For this review, we have identified several aspects (task design, subject preparation, actual scanning session, data processing, interpretation of results, and decision-making) where pediatric peculiarities should be taken into account. Further, we not only systematically identify common issues, but also provide solutions, based on our experience as well as a review of the pertinent literature. The aim is to provide the clinician as well as the imaging scientist with information that helps to plan, conduct, and interpret such a clinically-indicated exam in a way that maximizes benefit for, and minimizes the burden on the individual child.

Magnetoencephalographic Mapping of Epileptic Spike Population Using Distributed Source Analysis: Comparison With Intracranial Electroencephalographic Spikes

INTRODUCTION

This study evaluates magnetoencephalographic (MEG) spike population as compared with intracranial electroencephalographic (IEEG) spikes using a quantitative method based on distributed source analysis.

METHODS

We retrospectively studied eight patients with medically intractable epilepsy who had an MEG and subsequent IEEG monitoring. Fifty MEG spikes were analyzed in each patient using minimum norm estimate. For individual spikes, each vertex in the source space was considered activated when its source amplitude at the peak latency was higher than a threshold, which was set at 50% of the maximum amplitude over all vertices. We mapped the total count of activation at each vertex. We also analyzed 50 IEEG spikes in the same manner over the intracranial electrodes and created the activation count map. The location of the electrodes was obtained in the MEG source space by coregistering postimplantation computed tomography to MRI. We estimated the MEG- and IEEG-active regions associated with the spike populations using the vertices/electrodes with a count over 25.

RESULTS

The activation count maps of MEG spikes demonstrated the localization associated with the spike population by variable count values at each vertex. The MEG-active region overlapped with 65 to 85% of the IEEG-active region in our patient group.

CONCLUSIONS

Mapping the MEG spike population is valid for demonstrating the trend of spikes clustering in patients with epilepsy. In addition, comparison of MEG and IEEG spikes quantitatively may be informative for understanding their relationship.

Abnormal cortical activation during an auditory word comprehension task in benign childhood epilepsy with centrotemporal spikes: A magnetoencephalographic study

OBJECTIVE

Benign childhood epilepsy with centrotemporal spikes (BECTS), also known as rolandic epilepsy, has recently been reported to be associated with variable degrees of cognitive dysfunction. Many studies reported poor language ability in children with BECTS compared with healthy control children. To elucidate the harmful effects of BECTS on language cognition, we studied the magnetoencephalographic activity elicited by an auditory language comprehension task.

METHODS

The participants (N = 20) included 10 children diagnosed with BECTS (aged 10.8 ± 2.8 years) and 10 age-matched healthy children (control) (aged 10.6 ± 1.6 years). Cognitive function was assessed using general intellectual function and language ability. In patients with BECTS, we reviewed the clinical course and electroencephalogram (EEG) findings. We recorded the cortical responses elicited by an auditory language comprehension task using magnetoencephalography (MEG). We compared those results between groups and analyzed the correlation with cognitive scores and frequency of spikes.

RESULTS

The full-scale intelligence quotient (FSIQ) by the Wechsler Intelligence Scale for Children-4th edition was significantly reduced in the group with BECTS (96.4 ± 12.3) compared with the control group (110.0 ± 7.4). In half of the group with BECTS, the auditory comprehension score fell below the age-standard level. In the group with BECTS, the cortical activation during the task showed reduced intensity in language-associated areas such as the bilateral primary auditory cortex, left superior and mid-temporal areas, and inferior frontal area compared with those in the control group. In addition, the cortical activation in the left superior temporal area was negatively correlated with spike frequency and positively correlated with FSIQ in the group with BECTS. Conversely, the right inferior frontal and mid-temporal areas had increased the activations in the group with BECTS. From the time frequency analysis, low gamma band event-related desynchronization was reduced in the group with BECTS.

CONCLUSION

Epileptic spikes negatively influenced responsiveness to the auditory language comprehension task in the language-associated cortices. These findings suggest that epileptic spikes could have a negative impact on the functional activity in rolandic areas and become a reason to change the functional development of the language network.

Group Analysis in MNE-Python of Evoked Responses from a Tactile Stimulation Paradigm: A Pipeline for Reproducibility at Every Step of Processing, Going from Individual Sensor Space Representations to an across-Group Source Space Representation

An important aim of an analysis pipeline for magnetoencephalographic data is that it allows for the researcher spending maximal effort on making the statistical comparisons that will answer the questions of the researcher, while in turn spending minimal effort on the intricacies and machinery of the pipeline. I here present a set of functions and scripts that allow for setting up a clear, reproducible structure for separating raw and processed data into folders and files such that minimal effort can be spend on: (1) double-checking that the right input goes into the right functions; (2) making sure that output and intermediate steps can be accessed meaningfully; (3) applying operations efficiently across groups of subjects; (4) re-processing data if changes to any intermediate step are desirable. Applying the scripts requires only general knowledge about the Python language. The data analyses are neural responses to tactile stimulations of the right index finger in a group of 20 healthy participants acquired from an Elekta Neuromag System. Two analyses are presented: going from individual sensor space representations to, respectively, an across-group sensor space representation and an across-group source space representation. The processing steps covered for the first analysis are filtering the raw data, finding events of interest in the data, epoching data, finding and removing independent components related to eye blinks and heart beats, calculating participants' individual evoked responses by averaging over epoched data and calculating a grand average sensor space representation over participants. The second analysis starts from the participants' individual evoked responses and covers: estimating noise covariance, creating a forward model, creating an inverse operator, estimating distributed source activity on the cortical surface using a minimum norm procedure, morphing those estimates onto a common cortical template and calculating the patterns of activity that are statistically different from baseline. To estimate source activity, processing of the anatomy of subjects based on magnetic resonance imaging is necessary. The necessary steps are covered here: importing magnetic resonance images, segmenting the brain, estimating boundaries between different tissue layers, making fine-resolution scalp surfaces for facilitating co-registration, creating source spaces and creating volume conductors for each subject.