Towards an objective evaluation of EEG/MEG source estimation methods - The linear approach.

The spatial resolution of EEG/MEG source estimates, often described in terms of source leakage in the context of the inverse problem, poses constraints on the inferences that can be drawn from EEG/MEG source estimation results. Software packages for EEG/MEG data analysis offer a large choice of source estimation methods but few tools to experimental researchers for methods evaluation and comparison. Here, we describe a framework and tools for objective and intuitive resolution analysis of EEG/MEG source estimation based on linear systems analysis, and apply those to the most widely used distributed source estimation methods such as L2-minimum-norm estimation (L2-MNE) and linearly constrained minimum variance (LCMV) beamformers. Within this framework it is possible to define resolution metrics that define meaningful aspects of source estimation results (such as localization accuracy in terms of peak localization error, PLE, and spatial extent in terms of spatial deviation, SD) that are relevant to the task at hand and can easily be visualized. At the core of this framework is the resolution matrix, which describes the potential leakage from and into point sources (point-spread and cross-talk functions, or PSFs and CTFs, respectively). Importantly, for linear methods these functions allow generalizations to multiple sources or complex source distributions. This paper provides a tutorial-style introduction into linear EEG/MEG source estimation and resolution analysis aimed at experimental (rather than methods-oriented) researchers. We used this framework to demonstrate how L2-MNE-type as well as LCMV beamforming methods can be evaluated in practice using software tools that have only recently become available for routine use. Our novel methods comparison includes PLE and SD for a larger number of methods than in similar previous studies, such as unweighted, depth-weighted and normalized L2-MNE methods (including dSPM, sLORETA, eLORETA) and two LCMV beamformers. The results demonstrate that some methods can achieve low and even zero PLE for PSFs. However, their SD as well as both PLE and SD for CTFs are far less optimal for all methods, in particular for deep cortical areas. We hope that our paper will encourage EEG/MEG researchers to apply this approach to their own tasks at hand.

Spatial sampling of MEG and EEG based on generalized spatial-frequency analysis and optimal design.

In this paper, we analyze spatial sampling of electro- (EEG) and magnetoencephalography (MEG), where the electric or magnetic field is typically sampled on a curved surface such as the scalp. By simulating fields originating from a representative adult-male head, we study the spatial-frequency content in EEG as well as in on- and off-scalp MEG. This analysis suggests that on-scalp MEG, off-scalp MEG and EEG can benefit from up to 280, 90 and 110 spatial samples, respectively. In addition, we suggest a new approach to obtain sensor locations that are optimal with respect to prior assumptions. The approach also allows to control, e.g., the uniformity of the sensor locations. Based on our simulations, we argue that for a low number of spatial samples, model-informed non-uniform sampling can be beneficial. For a large number of samples, uniform sampling grids yield nearly the same total information as the model-informed grids.

Concurrent electrophysiological and hemodynamic measurements of evoked neural oscillations in human visual cortex using sparsely interleaved fast fMRI and EEG.

Electroencephalography (EEG) concurrently collected with functional magnetic resonance imaging (fMRI) is heavily distorted by the repetitive gradient coil switching during the fMRI acquisition. The performance of the typical template-based gradient artifact suppression method can be suboptimal because the artifact changes over time. Gradient artifact residuals also impede the subsequent suppression of ballistocardiography artifacts. Here we propose recording continuous EEG with temporally sparse fast fMRI (fast fMRI-EEG) to minimize the EEG artifacts caused by MRI gradient coil switching without significantly compromising the field-of-view and spatiotemporal resolution of fMRI. Using simultaneous multi-slice inverse imaging to achieve whole-brain fMRI with isotropic 5-mm resolution in 0.1 â€‹s, and performing these acquisitions once every 2 â€‹s, we have 95% of the duty cycle available to record EEG with substantially less gradient artifact. We found that the standard deviation of EEG signals over the entire acquisition period in fast fMRI-EEG was reduced to 54% of that in conventional concurrent echo-planar imaging (EPI) and EEG recordings (EPI-EEG) across participants. When measuring 15-Hz steady-state visual evoked potentials (SSVEPs), the baseline-normalized oscillatory neural response in fast fMRI-EEG was 2.5-fold of that in EPI-EEG. The functional MRI responses associated with the SSVEP delineated by EPI and fast fMRI were similar in the spatial distribution, the elicited waveform, and detection power. Sparsely interleaved fast fMRI-EEG provides high-quality EEG without substantially compromising the quality of fMRI in evoked response measurements, and has the potential utility for applications where the onset of the target stimulus cannot be precisely determined, such as epilepsy.

Comparison of beamformer implementations for MEG source localization.

Beamformers are applied for estimating spatiotemporal characteristics of neuronal sources underlying measured MEG/EEG signals. Several MEG analysis toolboxes include an implementation of a linearly constrained minimum-variance (LCMV) beamformer. However, differences in implementations and in their results complicate the selection and application of beamformers and may hinder their wider adoption in research and clinical use. Additionally, combinations of different MEG sensor types (such as magnetometers and planar gradiometers) and application of preprocessing methods for interference suppression, such as signal space separation (SSS), can affect the results in different ways for different implementations. So far, a systematic evaluation of the different implementations has not been performed. Here, we compared the localization performance of the LCMV beamformer pipelines in four widely used open-source toolboxes (MNE-Python, FieldTrip, DAiSS (SPM12), and Brainstorm) using datasets both with and without SSS interference suppression. We analyzed MEG data that were i) simulated, ii) recorded from a static and moving phantom, and iii) recorded from a healthy volunteer receiving auditory, visual, and somatosensory stimulation. We also investigated the effects of SSS and the combination of the magnetometer and gradiometer signals. We quantified how localization error and point-spread volume vary with the signal-to-noise ratio (SNR) in all four toolboxes. When applied carefully to MEG data with a typical SNR (3-15 â€‹dB), all four toolboxes localized the sources reliably; however, they differed in their sensitivity to preprocessing parameters. As expected, localizations were highly unreliable at very low SNR, but we found high localization error also at very high SNRs for the first three toolboxes while Brainstorm showed greater robustness but with lower spatial resolution. We also found that the SNR improvement offered by SSS led to more accurate localization.

Real-time computation of the TMS-induced electric field in a realistic head model.

Transcranial magnetic stimulation (TMS) is often targeted using a model of TMS-induced electric field (E). In such navigated TMS, the E-field models have been based on spherical approximation of the head. Such models omit the effects of cerebrospinal fluid (CSF) and gyral folding, leading to potentially large errors in the computed E-field. So far, realistic models have been too slow for interactive TMS navigation. We present computational methods that enable real-time solving of the E-field in a realistic five-compartment (5-C) head model that contains isotropic white matter, gray matter, CSF, skull and scalp. Using reciprocity and Geselowitz integral equation, we separate the computations to coil-dependent and -independent parts. For the Geselowitz integrals, we present a fast numerical quadrature. Further, we present a moment-matching approach for optimizing dipole-based coil models. We verified and benchmarked the new methods using simulations with over 100 coil locations. The new quadrature introduced a relative error (RE) of 0.3-0.6%. For a coil model with 42 dipoles, the total RE of the quadrature and coil model was 0.44-0.72%. Taking also other model errors into account, the contribution of the new approximations to the RE was 0.1%. For comparison, the RE due to omitting the separation of white and gray matter was >11%, and the RE due to omitting also the CSF was >23%. After the coil-independent part of the model has been built, E-fields can be computed very quickly: Using a standard PC and basic GPU, our solver computed the full E-field in a 5-C model in 9000 points on the cortex in 27 coil positions per second (cps). When the separation of white and gray matter was omitted, the speed was 43-65 cps. Solving only one component of the E-field tripled the speed. The presented methods enable real-time solving of the TMS-induced E-field in a realistic head model that contains the CSF and gyral folding. The new methodology allows more accurate targeting and precise adjustment of stimulation intensity during experimental or clinical TMS mapping.

Short-interval intracortical inhibition in human primary motor cortex: A multi-locus transcranial magnetic stimulation study.

Short-interval intracortical inhibition (SICI) has been studied with paired-pulse transcranial magnetic stimulation (TMS) by administering two pulses at a millisecond-scale interstimulus interval (ISI) to a single cortical target. It has, however, been difficult to study the interaction of nearby cortical targets with paired-pulse TMS. To overcome this limitation, we have developed a multi-locus TMS (mTMS) device, which allows controlling the stimulus location electronically. Here, we applied mTMS to study SICI in primary motor cortex with paired pulses targeted to adjacent locations, aiming to quantify the extent of the cortical region producing SICI in the location of a test stimulus. We varied the location and timing of the conditioning stimulus with respect to a test stimulus targeted to the cortical hotspot of the abductor pollicis brevis (APB) in order to study their effects on motor evoked potentials. We further applied a two-coil protocol with the conditioning stimulus given by an oval coil only to the surroundings of the APB hotspot, to which a subsequent test stimulus was administered with a figure-of-eight coil. The strongest SICI occurred at ISIs below 1 ms and at ISIs around 2.5 ms. These ISIs increased when the conditioning stimulus receded from the APB hotspot. Our two-coil paired-pulse TMS study suggests that SICI at ISIs of 0.5 and 2.5 ms originate from different mechanisms or neuronal elements.

The impact of improved MEG-MRI co-registration on MEG connectivity analysis.

Co-registration between structural head images and functional MEG data is needed for anatomically-informed MEG data analysis. Despite the efforts to minimize the co-registration error, conventional landmark- and surface-based strategies for co-registering head and MEG device coordinates achieve an accuracy of typically 5-10 mm. Recent advances in instrumentation and technical solutions, such as the development of hybrid ultra-low-field (ULF) MRI-MEG devices or the use of 3D-printed individualized foam head-casts, promise unprecedented co-registration accuracy, i.e., 2 mm or better. In the present study, we assess through simulations the impact of such an improved co-registration on MEG connectivity analysis. We generated synthetic MEG recordings for pairs of connected cortical sources with variable locations. We then assessed the capability to reconstruct source-level connectivity from these recordings for 0-15-mm co-registration error, three levels of head modeling detail (one-, three- and four-compartment models), two source estimation techniques (linearly constrained minimum-variance beamforming and minimum-norm estimation MNE) and five separate connectivity metrics (imaginary coherency, phase-locking value, amplitude-envelope correlation, phase-slope index and frequency-domain Granger causality). We found that beamforming can better take advantage of an accurate co-registration than MNE. Specifically, when the co-registration error was smaller than 3 mm, the relative error in connectivity estimates was down to one-third of that observed with typical co-registration errors. MNE provided stable results for a wide range of co-registration errors, while the performance of beamforming rapidly degraded as the co-registration error increased. Furthermore, we found that even moderate co-registration errors (>6 mm, on average) essentially decrease the difference of four- and three- or one-compartment models. Hence, a precise co-registration is important if one wants to take full advantage of highly accurate head models for connectivity analysis. We conclude that an improved co-registration will be beneficial for reliable connectivity analysis and effective use of highly accurate head models in future MEG connectivity studies.

Truncated RAP-MUSIC (TRAP-MUSIC) for MEG and EEG source localization.

Electrically active brain regions can be located applying MUltiple SIgnal Classification (MUSIC) on magneto- or electroencephalographic (MEG; EEG) data. We introduce a new MUSIC method, called truncated recursively-applied-and-projected MUSIC (TRAP-MUSIC). It corrects a hidden deficiency of the conventional RAP-MUSIC algorithm, which prevents estimation of the true number of brain-signal sources accurately. The correction is done by applying a sequential dimension reduction to the signal-subspace projection. We show that TRAP-MUSIC significantly improves the performance of MUSIC-type localization; in particular, it successfully and robustly locates active brain regions and estimates their number. We compare TRAP-MUSIC and RAP-MUSIC in simulations with varying key parameters, e.g., signal-to-noise ratio, correlation between source time-courses, and initial estimate for the dimension of the signal space. In addition, we validate TRAP-MUSIC with measured MEG data. We suggest that with the proposed TRAP-MUSIC method, MUSIC-type localization could become more reliable and suitable for various online and offline MEG and EEG applications.

Requirements for Coregistration Accuracy in On-Scalp MEG.

Recent advances in magnetic sensing has made on-scalp magnetoencephalography (MEG) possible. In particular, optically-pumped magnetometers (OPMs) have reached sensitivity levels that enable their use in MEG. In contrast to the SQUID sensors used in current MEG systems, OPMs do not require cryogenic cooling and can thus be placed within millimetres from the head, enabling the construction of sensor arrays that conform to the shape of an individual's head. To properly estimate the location of neural sources within the brain, one must accurately know the position and orientation of sensors in relation to the head. With the adaptable on-scalp MEG sensor arrays, this coregistration becomes more challenging than in current SQUID-based MEG systems that use rigid sensor arrays. Here, we used simulations to quantify how accurately one needs to know the position and orientation of sensors in an on-scalp MEG system. The effects that different types of localisation errors have on forward modelling and source estimates obtained by minimum-norm estimation, dipole fitting, and beamforming are detailed. We found that sensor position errors generally have a larger effect than orientation errors and that these errors affect the localisation accuracy of superficial sources the most. To obtain similar or higher accuracy than with current SQUID-based MEG systems, RMS sensor position and orientation errors should be [Formula: see text] and [Formula: see text], respectively.

Measuring MEG closer to the brain: Performance of on-scalp sensor arrays.

Optically-pumped magnetometers (OPMs) have recently reached sensitivity levels required for magnetoencephalography (MEG). OPMs do not need cryogenics and can thus be placed within millimetres from the scalp into an array that adapts to the individual head size and shape, thereby reducing the distance from cortical sources to the sensors. Here, we quantified the improvement in recording MEG with hypothetical on-scalp OPM arrays compared to a 306-channel state-of-the-art SQUID array (102 magnetometers and 204 planar gradiometers). We simulated OPM arrays that measured either normal (nOPM; 102 sensors), tangential (tOPM; 204 sensors), or all components (aOPM; 306 sensors) of the magnetic field. We built forward models based on magnetic resonance images of 10 adult heads; we employed a three-compartment boundary element model and distributed current dipoles evenly across the cortical mantle. Compared to the SQUID magnetometers, nOPM and tOPM yielded 7.5 and 5.3 times higher signal power, while the correlations between the field patterns of source dipoles were reduced by factors of 2.8 and 3.6, respectively. Values of the field-pattern correlations were similar across nOPM, tOPM and SQUID gradiometers. Volume currents reduced the signals of primary currents on average by 10%, 72% and 15% in nOPM, tOPM and SQUID magnetometers, respectively. The information capacities of the OPM arrays were clearly higher than that of the SQUID array. The dipole-localization accuracies of the arrays were similar while the minimum-norm-based point-spread functions were on average 2.4 and 2.5 times more spread for the SQUID array compared to nOPM and tOPM arrays, respectively.

Recovering TMS-evoked EEG responses masked by muscle artifacts.

Combined transcranial magnetic stimulation (TMS) and electroencephalography (EEG) often suffers from large muscle artifacts. Muscle artifacts can be removed using signal-space projection (SSP), but this can make the visual interpretation of the remaining EEG data difficult. We suggest to use an additional step after SSP that we call source-informed reconstruction (SIR). SSP-SIR improves substantially the signal quality of artifactual TMS-EEG data, causing minimal distortion in the neuronal signal components. In the SSP-SIR approach, we first project out the muscle artifact using SSP. Utilizing an anatomical model and the remaining signal, we estimate an equivalent source distribution in the brain. Finally, we map the obtained source estimate onto the original signal space, again using anatomical information. This approach restores the neuronal signals in the sensor space and interpolates EEG traces onto the completely rejected channels. The introduced algorithm efficiently suppresses TMS-related muscle artifacts in EEG while retaining well the neuronal EEG topographies and signals. With the presented method, we can remove muscle artifacts from TMS-EEG data and recover the underlying brain responses without compromising the readability of the signals of interest.

Assessment of myocardial infarct size with body surface potential mapping: validation against contrast-enhanced cardiac magnetic resonance imaging.

BACKGROUND: Assessment of myocardial infarct (MI) size is important for therapeutic and prognostic reasons. We used body surface potential mapping (BSPM) to evaluate whether single-lead electrocardiographic variables can assess MI size. METHODS: We performed BSPM with 120 leads covering the front and back chest (plus limb leads) on 57 patients at different phases of MI: acutely, during healing, and in the chronic phase. Final MI size was determined by contrast-enhanced cardiac magnetic resonance imaging (DE-CMR) and correlated with various computed depolarization- and repolarization-phase BSPM variables. We also calculated correlations between BSPM variables and enzymatic MI size (peak CK-MBm). RESULTS: BSPM variables reflecting the Q- and R wave showed strong correlations with MI size at all stages of MI. R width performed the best, showing its strongest correlation with MI size on the upper right back, there representing the width of the "reciprocal Q wave" (r = 0.64-0.71 for DE-CMR, r = 0.57-0.64 for CK-MBm, P < 0.0001). Repolarization-phase variables showed only weak correlations with MI size in the acute phase, but these correlations improved during MI healing. T-wave variables and the QRSSTT integral showed their best correlations with DE-CMR defined MI size on the precordial area, at best r = -0.57, P < 0.0001 in the chronic phase. The best performing BSPM variables could differentiate between large and small infarcts at all stages of MI. CONCLUSIONS: Computed, single-lead electrocardiographic variables can estimate the final infarct size at all stages of MI, and differentiate large infarcts from small.

Comparison of three-shell and simplified volume conductor models in magnetoencephalography.

Experimental MEG source imaging studies have typically been carried out with either a spherically symmetric head model or a single-shell boundary-element (BEM) model that is shaped according to the inner skull surface. The concepts and comparisons behind these simplified models have led to misunderstandings regarding the role of skull and scalp in MEG. In this work, we assess the forward-model errors due to different skull/scalp approximations and due to differences and errors in model geometries. We built five anatomical models of a volunteer using a set of T1-weighted MR scans and three common toolboxes. Three of the models represented typical models in experimental MEG, one was manually constructed, and one contained a major segmentation error at the skull base. For these anatomical models, we built forward models using four simplified approaches and a three-shell BEM approach that has been used as reference in previous studies. Our reference model contained in addition the skull fine-structure (spongy bone). We computed signal topographies for cortically constrained sources in the left hemisphere and compared the topographies using relative error and correlation metrics. The results show that the spongy bone has a minimal effect on MEG topographies, and thus the skull approximation of the three-shell model is justified. The three-shell model performed best, followed by the corrected-sphere and single-shell models, whereas the local-spheres and single-sphere models were clearly worse. The three-shell model was the most robust against the introduced segmentation error. In contrast to earlier claims, there was no noteworthy difference in the computation times between the realistically-shaped and sphere-based models, and the manual effort of building a three-shell model and a simplified model is comparable. We thus recommend the realistically-shaped three-shell model for experimental MEG work. In cases where this is not possible, we recommend a realistically-shaped corrected-sphere or single-shell model.

A framework for the design of flexible cross-talk functions for spatial filtering of EEG/MEG data: DeFleCT.

Brain activation estimated from EEG and MEG data is the basis for a number of time-series analyses. In these applications, it is essential to minimize "leakage" or "cross-talk" of the estimates among brain areas. Here, we present a novel framework that allows the design of flexible cross-talk functions (DeFleCT), combining three types of constraints: (1) full separation of multiple discrete brain sources, (2) minimization of contributions from other (distributed) brain sources, and (3) minimization of the contribution from measurement noise. Our framework allows the design of novel estimators by combining knowledge about discrete sources with constraints on distributed source activity and knowledge about noise covariance. These estimators will be useful in situations where assumptions about sources of interest need to be combined with uncertain information about additional sources that may contaminate the signal (e.g. distributed sources), and for which existing methods may not yield optimal solutions. We also show how existing estimators, such as maximum-likelihood dipole estimation, L2 minimum-norm estimation, and linearly-constrained minimum variance as well as null-beamformers, can be derived as special cases from this general formalism. The performance of the resulting estimators is demonstrated for the estimation of discrete sources and regions-of-interest in simulations of combined EEG/MEG data. Our framework will be useful for EEG/MEG studies applying time-series analysis in source space as well as for the evaluation and comparison of linear estimators.

Minimum-norm cortical source estimation in layered head models is robust against skull conductivity error.

The conductivity profile of the head has a major effect on EEG signals, but unfortunately the conductivity for the most important compartment, skull, is only poorly known. In dipole modeling studies, errors in modeled skull conductivity have been considered to have a detrimental effect on EEG source estimation. However, as dipole models are very restrictive, those results cannot be generalized to other source estimation methods. In this work, we studied the sensitivity of EEG and combined MEG+EEG source estimation to errors in skull conductivity using a distributed source model and minimum-norm (MN) estimation. We used a MEG/EEG modeling set-up that reflected state-of-the-art practices of experimental research. Cortical surfaces were segmented and realistically-shaped three-layer anatomical head models were constructed, and forward models were built with Galerkin boundary element method while varying the skull conductivity. Lead-field topographies and MN spatial filter vectors were compared across conductivities, and the localization and spatial spread of the MN estimators were assessed using intuitive resolution metrics. The results showed that the MN estimator is robust against errors in skull conductivity: the conductivity had a moderate effect on amplitudes of lead fields and spatial filter vectors, but the effect on corresponding morphologies was small. The localization performance of the EEG or combined MEG+EEG MN estimator was only minimally affected by the conductivity error, while the spread of the estimate varied slightly. Thus, the uncertainty with respect to skull conductivity should not prevent researchers from applying minimum norm estimation to EEG or combined MEG+EEG data. Comparing our results to those obtained earlier with dipole models shows that general judgment on the performance of an imaging modality should not be based on analysis with one source estimation method only.

Predicting recovery of myocardial function by electrocardiography after acute infarction.

BACKGROUND: In acute ischemic left ventricular (LV) dysfunction, distinguishing viable myocardium is clinically important. METHODS: Body surface potential mapping (Electrocardiography [ECG] with 123 leads), was recorded in 62 patients with acute coronary syndrome (ACS). ECG variables were computed from de- and repolarization phases. LV segmental wall motion was assessed by echocardiography acutely and after 1 year. RESULTS: The number of dysfunctional segments (DFS) diminished during follow-up in 37 patients (recovery group) and remained the same or increased in 25 patients (nonrecovery group). Acutely, DFS was 5.7 ± 2.1 versus 4.4 ± 2.4 (P = 0.02), and peak CK-MBm 141 ± 157 versus 156 ± 167 µg/L (P = 0.78) in the recovery versus nonrecovery group. At follow-up, DFS was 1.9 ± 1.7 versus 6.5 ± 2.6 (P < 0.001). The best ECG variable to predict decrease in DFS depended on the region of acute LV dysfunction: The best variable in the left anterior descending region was the integral of the first QRS integral (area under the curve [AUC] 0.82, P = 0.002); in the right coronary artery region, this was the integral of the ST segment (AUC 0.98, P = 0.003); and in the left circumflex region, the area including the ST segment and the T wave (AUC 0.97, P = 0.006). CONCLUSIONS: In ACS patients, computed ECG variables predict recovery of LV function from ischemic myocardial injury, even in the presence of comparable CK-MBm release and LV dysfunction.

Prefrontal theta phase-dependent rTMS-induced plasticity of cortical and behavioral responses in human cortex.

BACKGROUND: Prefrontal theta oscillations are involved in neuronal information transfer and retention. Phases along the theta cycle represent varied excitability states, whereby high-excitability states correspond to high-frequency neuronal activity and heightened capacity for plasticity induction, as demonstrated in animal studies. Human studies corroborate this model and suggest a core role of prefrontal theta activity in working memory (WM). OBJECTIVE/HYPOTHESIS: We aimed at modulating prefrontal neuronal excitability and WM performance in healthy humans, using real-time EEG analysis for triggering repetitive transcranial magnetic stimulation (rTMS) theta-phase synchronized to the left dorsomedial prefrontal cortex. METHODS: 16 subjects underwent 3 different rTMS interventions on separate days, with pulses triggered according to the individual's real-time EEG activity: 400 rTMS gamma-frequency (100 Hz) triplet bursts applied during either the negative peak of the prefrontal theta oscillation, the positive peak, or at random phase. Changes in cortical excitability were assessed with EEG responses following single-pulse TMS, and behavioral effects by using a WM task. RESULTS: Negative-peak rTMS increased single-pulse TMS-induced prefrontal theta power and theta-gamma phase-amplitude coupling, and decreased WM response time. In contrast, positive-peak rTMS decreased prefrontal theta power, while no changes were observed after random-phase rTMS. CONCLUSION: Findings point to the feasibility of EEG-TMS technology in a theta-gamma phase-amplitude coupling mode for effectively modifying WM networks in human prefrontal cortex, with potential for therapeutic applications.

µ-rhythm phase from somatosensory but not motor cortex correlates with corticospinal excitability in EEG-triggered TMS.

BACKGROUND: Sensorimotor µ-rhythm phase is correlated with corticospinal excitability. Transcranial magnetic stimulation (TMS) of motor cortex results in larger motor evoked potentials (MEPs) during the negative peak of the EEG oscillation as extracted with a surface Laplacian. However, the anatomical source of the relevant oscillation is not clear and demonstration of the relationship is sensitive to the choice of EEG montage. OBJECTIVE/HYPOTHESIS: Here, we compared two EEG montages preferentially sensitive to oscillations originating from the crown of precentral gyrus (dorsal premotor cortex) vs. postcentral gyrus (secondary somatosensory cortex). We hypothesized that the EEG signal from precentral gyrus would correlate more strongly with MEP amplitude, given that the corticospinal neurons are located in the anterior wall of the sulcus and the corticospinal tract has input from premotor cortex. NEW METHOD: Real-time EEG-triggered TMS of motor cortex was applied in 6 different conditions in randomly interleaved order, 3 phase conditions (positive peak, negative peak, random phase of the ongoing µ-oscillation), and each phase condition for 2 different EEG montages corresponding to oscillations preferentially originating in precentral gyrus (premotor cortex) vs. postcentral gyrus (somatosensory cortex), extracted using FCC3h vs. C3 centered EEG montages. RESULTS: The negative vs. positive peak of sensorimotor µ-rhythm as extracted from the C3 montage (postcentral gyrus, somatosensory cortex) correlated with states of high vs. low corticospinal excitability (p < 0.001), replicating previous findings. However, no significant correlation was found for sensorimotor µ-rhythm as extracted from the neighboring FCC3 montage (precentral gyrus, premotor cortex). This implies that EEG-signals from the somatosensory cortex are better predictors of corticospinal excitability than EEG-signals from the motor areas. CONCLUSIONS: The extraction of a brain oscillation whose phase corresponds to corticospinal excitability is highly sensitive to the selected EEG montage and the location of the EEG sensors on the scalp. Here, the cortical source of EEG oscillations predicting response amplitude does not correspond to the cortical target of the stimulation, indicating that even in this simple case, a specific neuronal pathway from somatosensory cortex to primary motor cortex is involved.

Closed-loop optimization of transcranial magnetic stimulation with electroencephalography feedback.

BACKGROUND: Transcranial magnetic stimulation (TMS) is widely used in brain research and treatment of various brain dysfunctions. However, the optimal way to target stimulation and administer TMS therapies, for example, where and in which electric field direction the stimuli should be given, is yet to be determined. OBJECTIVE: To develop an automated closed-loop system for adjusting TMS parameters (in this work, the stimulus orientation) online based on TMS-evoked brain activity measured with electroencephalography (EEG). METHODS: We developed an automated closed-loop TMS-EEG set-up. In this set-up, the stimulus parameters are electronically adjusted with multi-locus TMS. As a proof of concept, we developed an algorithm that automatically optimizes the stimulation orientation based on single-trial EEG responses. We applied the algorithm to determine the electric field orientation that maximizes the amplitude of the TMS-EEG responses. The validation of the algorithm was performed with six healthy volunteers, repeating the search twenty times for each subject. RESULTS: The validation demonstrated that the closed-loop control worked as desired despite the large variation in the single-trial EEG responses. We were often able to get close to the orientation that maximizes the EEG amplitude with only a few tens of pulses. CONCLUSION: Optimizing stimulation with EEG feedback in a closed-loop manner is feasible and enables effective coupling to brain activity.

Accuracy and precision of navigated transcranial magnetic stimulation.

Objective.Transcranial magnetic stimulation (TMS) induces an electric field (E-field) in the cortex. To facilitate stimulation targeting, image-guided neuronavigation systems have been introduced. Such systems track the placement of the coil with respect to the head and visualize the estimated cortical stimulation location on an anatomical brain image in real time. The accuracy and precision of the neuronavigation is affected by multiple factors. Our aim was to analyze how different factors in TMS neuronavigation affect the accuracy and precision of the coil-head coregistration and the estimated E-field.Approach.By performing simulations, we estimated navigation errors due to distortions in magnetic resonance images (MRIs), head-to-MRI registration (landmark- and surface-based registrations), localization and movement of the head tracker, and localization of the coil tracker. We analyzed the effect of these errors on coil and head coregistration and on the induced E-field as determined with simplistic and realistic head models.Main results.Average total coregistration accuracies were in the range of 2.2-3.6 mm and 1°; precision values were about half of the accuracy values. The coregistration errors were mainly due to head-to-MRI registration with average accuracies 1.5-1.9 mm/0.2-0.4° and precisions 0.5-0.8 mm/0.1-0.2° better with surface-based registration. The other major source of error was the movement of the head tracker with average accuracy of 1.5 mm and precision of 1.1 mm. When assessed within an E-field method, the average accuracies of the peak E-field location, orientation, and magnitude ranged between 1.5 and 5.0 mm, 0.9 and 4.8°, and 4.4 and 8.5% across the E-field models studied. The largest errors were obtained with the landmark-based registration. When computing another accuracy measure with the most realistic E-field model as a reference, the accuracies tended to improve from about 10 mm/15°/25% to about 2 mm/2°/5% when increasing realism of the E-field model.Significance.The results of this comprehensive analysis help TMS operators to recognize the main sources of error in TMS navigation and that the coregistration errors and their effect in the E-field estimation depend on the methods applied. To ensure reliable TMS navigation, we recommend surface-based head-to-MRI registration and realistic models for E-field computations.