No file available
To read the file of this research,
you can request a copy directly from the authors.
Comparison of EEG source localization using simplified and anatomically accurate head models in younger and older adults
Abstract
p>Accuracy of electroencephalography (EEG) source localization relies on the volume conduction head model. A previous analysis of young adults has shown that simplified head models have larger source localization errors when compared with head models based on magnetic resonance images (MRIs). As obtaining individual MRIs may not always be feasible, researchers often use generic head models based on template MRIs. It is unclear how much error would be introduced using template MRI head models in older adults that likely have differences in brain structure compared to young adults. The primary goal of this study was to determine the error caused by using simplified head models without individual-specific MRIs in both younger and older adults. We collected high-density EEG during uneven terrain walking and motor imagery for 15 younger (22±3 years) and 21 older adults (74±5 years) and obtained T1-weighted MRI for each individual. We performed equivalent dipole fitting after independent component analysis to obtain brain source locations using four forward modeling pipelines with increasing complexity. These pipelines included: 1) a generic head model with template electrode positions or 2) digitized electrode positions, 3) individual-specific head models with digitized electrode positions using simplified tissue segmentation, or 4) anatomically accurate segmentation. We found that when compared to the anatomically accurate individual-specific head models, performing dipole fitting with generic head models led to similar source localization discrepancies (up to 2 cm) for younger and older adults. Co-registering digitized electrode locations to the generic head models reduced source localization discrepancies by ~6 mm. Additionally, we found that source depths generally increased with skull conductivity for the representative young adult but not as much for the older adult. Our results can help inform a more accurate interpretation of brain areas in EEG studies when individual MRIs are unavailable. </p
... Although we did not construct individual head models for each participant, we did use subject-specific MRI images to visually evaluate the alignment between the MRI and the head model. Using subject specific models would have increased source localization accuracy, especially if the head model included realistic tissue conductivity values [62]. Nonetheless, our analysis was able to identify sources comparable to previous studies that recorded EEG during treadmill and balance beam-walking [2], [21], [63], [64]. ...
There is a need to develop appropriate balance training interventions to minimize the risk of falls. Recently, we found that intermittent visual occlusions can substantially improve the effectiveness and retention of balance beam walking practice (Symeonidou & Ferris, 2022). We sought to determine how the intermittent visual occlusions affect electrocortical activity during beam walking. We hypothesized that areas involved in sensorimotor processing and balance control would demonstrate spectral power changes and inter-trial coherence modulations after loss and restoration of vision. Ten healthy young adults practiced walking on a treadmill-mounted balance beam while wearing high-density EEG and experiencing reoccurring visual occlusions. Results revealed spectral power fluctuations and inter-trial coherence changes in the visual, occipital, temporal, and sensorimotor cortex as well as the posterior parietal cortex and the anterior cingulate. We observed a prolonged alpha increase in the occipital, temporal, sensorimotor, and posterior parietal cortex after the occlusion onset. In contrast, the anterior cingulate showed a strong alpha and theta increase after the occlusion offset. We observed transient phase synchrony in the alpha, theta, and beta bands within the sensory, posterior parietal, and anterior cingulate cortices immediately after occlusion onset and offset. Intermittent visual occlusions induced electrocortical spectral power and inter-trial coherence changes in a wide range of frequencies within cortical areas relevant for multisensory integration and processing as well as balance control. Our training intervention could be implemented in senior and rehabilitation centers, improving the quality of life of elderly and neurologically impaired individuals.
We developed a method for altering terrain unevenness on a treadmill to study gait kinematics. Terrain consisted of rigid polyurethane disks (12.7 cm diameter, 1.3–3.8 cm tall) which attached to the treadmill belt using hook-and-loop fasteners. Here, we tested four terrain unevenness conditions: Flat, Low, Medium, and High. The main objective was to test the hypothesis that increasing the unevenness of the terrain would result in greater gait kinematic variability. Seventeen younger adults (age 20–40 years), 25 higher-functioning older adults (age 65+ years), and 29 lower-functioning older adults (age 65+ years, Short Physical Performance Battery score < 10) participated. We customized the treadmill speed to each participant’s walking ability, keeping the speed constant across all four terrain conditions. Participants completed two 3-minute walking trials per condition. Using an inertial measurement unit placed over the sacrum and pressure sensors in the shoes, we calculated the stride-to-stride variability in step duration and sacral excursion (coefficient of variation; standard deviation expressed as percentage of the mean). Participants also self-reported their perceived stability for each condition. Terrain was a significant predictor of step duration variability, which roughly doubled from Flat to High terrain for all participant groups: younger adults (Flat 4.0%, High 8.2%), higher-functioning older adults (Flat 5.0%, High 8.9%), lower-functioning older adults (Flat 7.0%, High 14.1%). Similarly, all groups exhibited significant increases in sacral excursion variability for the Medium and High uneven terrain conditions, compared to Flat. Participants were also significantly more likely to report feeling less stable walking over all three uneven terrain conditions compared to Flat. These findings support the hypothesis that altering terrain unevenness on a treadmill will increase gait kinematic variability and reduce perceived stability in younger and older adults.
Objective. Magneto- and electroencephalography (M/EEG) measurements record a mix of signals from the brain, eyes, and muscles. These signals can be disentangled for artifact cleaning e.g. using spatial filtering techniques. However, correctly localizing and identifying these components relies on head models that so far only take brain sources into account. Approach. We thus developed the Head Artifact Model using Tripoles (HArtMuT). This volume conduction head model extends to the neck and includes brain sources as well as sources representing eyes and muscles that can be modeled as single dipoles, symmetrical dipoles, and tripoles. We compared a HArtMuT four-layer boundary element model (BEM) with the EEGLAB standard head model on their localization accuracy and residual variance (RV) using a HArtMuT finite element model (FEM) as ground truth. We also evaluated the RV on real-world data of mobile participants, comparing different HArtMuT BEM types with the EEGLAB standard head model. Main results. We found that HArtMuT improves localization for all sources, especially non-brain, and localization error and RV of non-brain sources were in the same range as those of brain sources. The best results were achieved by using cortical dipoles, muscular tripoles, and ocular symmetric dipoles, but dipolar sources alone can already lead to convincing results. Significance. We conclude that HArtMuT is well suited for modeling eye and muscle contributions to the M/EEG signal. It can be used to localize sources and to identify brain, eye, and muscle components. HArtMuT is freely available and can be integrated into standard software.
Researchers can improve the ecological validity of brain research by studying humans moving in real-world settings. Recent work shows that dual-layer EEG can improve the fidelity of electrocortical recordings during gait, but it is unclear whether these positive results extrapolate to non-locomotor paradigms. For our study, we recorded brain activity with dual-layer EEG while participants played table tennis, a whole-body, responsive sport that could help investigate visuomotor feedback, object interception, and performance monitoring. We characterized artifacts with time-frequency analyses and correlated scalp and reference noise data to determine how well different sensors captured artifacts. As expected, individual scalp channels correlated more with noise-matched channel time series than with head and body acceleration. We then compared artifact removal methods with and without the use of the dual-layer noise electrodes. Independent Component Analysis separated channels into components, and we counted the number of high-quality brain components based on the fit of a dipole model and using an automated labeling algorithm. We found that using noise electrodes for data processing provided cleaner brain components. These results advance technological approaches for recording high fidelity brain dynamics in human behaviors requiring whole body movement, which will be useful for brain science research.
Almost 25% of all older adults experience difficulty walking. Mobility difficulties for older adults are more pronounced when they perform a simultaneous cognitive task while walking (i.e., dual task walking). Although it is known that aging results in widespread brain atrophy, few studies have integrated across more than one neuroimaging modality to comprehensively examine the structural neural correlates that may underlie dual task walking in older age. We collected spatiotemporal gait data during single and dual task walking for 37 young (18–34 years) and 23 older adults (66–86 years). We also collected T1-weighted and diffusion-weighted MRI scans to determine how brain structure differs in older age and relates to dual task walking. We addressed two aims: (1) to characterize age differences in brain structure across a range of metrics including volumetric, surface, and white matter microstructure; and (2) to test for age group differences in the relationship between brain structure and the dual task cost (DTcost) of gait speed and variability. Key findings included widespread brain atrophy for the older adults, with the most pronounced age differences in brain regions related to sensorimotor processing. We also found multiple associations between regional brain atrophy and greater DTcost of gait speed and variability for the older adults. The older adults showed a relationship of both thinner temporal cortex and shallower sulcal depth in the frontal, sensorimotor, and parietal cortices with greater DTcost of gait. Additionally, the older adults showed a relationship of ventricular volume and superior longitudinal fasciculus free-water corrected axial and radial diffusivity with greater DTcost of gait. These relationships were not present for the young adults. Stepwise multiple regression found sulcal depth in the left precentral gyrus, axial diffusivity in the superior longitudinal fasciculus, and sex to best predict DTcost of gait speed, and cortical thickness in the superior temporal gyrus to best predict DTcost of gait variability for older adults. These results contribute to scientific understanding of how individual variations in brain structure are associated with mobility function in aging. This has implications for uncovering mechanisms of brain aging and for identifying target regions for mobility interventions for aging populations.
The parallel execution of two motor tasks can lead to performance decrements in either one or both of the tasks. Age-related declines can further magnify the underlying competition for cognitive resources. However, little is known about the neural dynamics underlying motor resource allocation during dual-task walking. To better understand motor resource conflicts, this study investigated sensorimotor brain rhythms in younger and older adults using a dual-task protocol. Time-frequency data from two independent component motor clusters were extracted from electroencephalography data during sitting and walking with an additional task requiring manual responses. Button press-related desynchronization in the alpha and beta frequency range were analyzed for the impact of age (< 35 years, ≥ 70 years) and motor task (sitting, walking). Button press-related desynchronization in the beta band was more pronounced for older participants and both age groups demonstrated less pronounced desynchronizations in both frequency bands during walking compared to sitting. Older participants revealed less power modulations between sitting and walking, and less pronounced changes in beta and alpha suppression were associated with greater slowing in walking speed. Our results indicate age-specific allocations strategies during dual-task walking as well as interdependencies of concurrently performed motor tasks reflected in modulations of sensorimotor rhythms.
Electroencephalogram (EEG) power reductions in the aging brain have been described by numerous previous studies. However, the underlying mechanism for the observed brain signal power reduction remains unclear. One possible cause for reduced EEG signals in elderly subjects might be the increased distance from the primary neural electrical currents on the cortex to the scalp electrodes as the result of cortical atrophies. While brain shrinkage itself reflects age-related neurological changes, the effects of changes in the distribution of electrical conductivity are often not distinguished from altered neural activity when interpreting EEG power reductions. To address this ambiguity, we employed EEG forward models to investigate whether brain shrinkage is a major factor for the signal attenuation in the aging brain. We simulated brain shrinkage in spherical and realistic brain models and found that changes in the conductor geometry cannot fully account for the EEG power reductions even when the brain was shrunk to unrealistic sizes. Our results quantify the extent of power reductions from brain shrinkage and pave the way for more accurate inferences about deficient neural activity and circuit integrity based on EEG power reductions in the aging population.
Magneto-/Electro-encephalography (M/EEG) commonly uses (fast) Fourier transformation to compute power spectral density (PSD). However, the resulting PSD plot lacks temporal information, making interpretation sometimes equivocal. For example, consider two different PSDs: a central parietal EEG PSD with twin peaks at 10 Hz and 20 Hz and a central parietal PSD with twin peaks at 10 Hz and 50 Hz. We can assume the first PSD shows a mu rhythm and the second harmonic; however, the latter PSD likely shows an alpha peak and an independent line noise. Without prior knowledge, however, the PSD alone cannot distinguish between the two cases. To address this limitation of PSD, we propose using cross-frequency power–power coupling (PPC) as a post-processing of independent component (IC) analysis (ICA) to distinguish brain components from muscle and environmental artifact sources. We conclude that post-ICA PPC analysis could serve as a new data-driven EEG classifier in M/EEG studies. For the reader’s convenience, we offer a brief literature overview on the disparate use of PPC. The proposed cross-frequency power–power coupling analysis toolbox (PowPowCAT) is a free, open-source toolbox, which works as an EEGLAB extension.
Behavioral findings suggest that aging alters the involvement of cortical sensorimotor mechanisms in postural control. However, corresponding accounts of the underlying neural mechanisms remain sparse, especially the extent to which these mechanisms are affected during more demanding tasks. Here, we set out to elucidate cortical correlates of altered postural stability in younger and older adults. 3D body motion tracking and high‐density electroencephalography (EEG) were measured while 14 young adults (mean age = 24 years, 43% women) and 14 older adults (mean age = 77 years, 50% women) performed a continuous balance task under four different conditions. Manipulations were applied to the base of support (either regular or tandem (heel‐to‐toe) stance) and visual input (either static visual field or dynamic optic flow). Standing in tandem, the more challenging position, resulted in increased sway for both age groups, but for the older adults, only this effect was exacerbated when combined with optic flow compared to the static visual display. These changes in stability were accompanied by neuro‐oscillatory modulations localized to midfrontal and parietal regions. A cluster of electro‐cortical sources localized to the supplementary motor area showed a large increase in theta spectral power (4–7 Hz) during tandem stance, and this modulation was much more pronounced for the younger group. Additionally, the older group displayed widespread mu (8–12 Hz) and beta (13–30 Hz) suppression as balance tasks placed more demands on postural control, especially during tandem stance. These findings may have substantial utility in identifying early cortical correlates of balance impairments in otherwise healthy older adults.
Skull conductivity has a substantial influence on EEG and combined EEG and MEG source analysis as well as on optimized transcranial electric stimulation. To overcome the use of standard literature values, we propose a non-invasive two-level calibration procedure to estimate skull conductivity individually in a group study with twenty healthy adults. Our procedure requires only an additional run of combined somatosensory evoked potential and field data, which can be easily integrated in EEG/MEG experiments. The calibration procedure uses the P20/N20 topographies and subject-specific realistic head models from MRI. We investigate the inter-subject variability of skull conductivity and relate it to skull thickness, age and gender of the subjects, to the individual scalp P20/N20 surface distance between the P20 potential peak and the N20 potential trough as well as to the individual source depth of the P20/N20 source. We found a considerable inter-subject variability for (calibrated) skull conductivity (8.44 ± 4.84 mS/m) and skull thickness (5.97 ± 1.19 mm) with a statistically significant correlation between them (rho = 0.52). Age showed a statistically significant negative correlation with skull conductivity (rho =-0.5). Furthermore, P20/N20 surface distance and source depth showed large inter-subject variability of 12.08 ± 3.21 cm and 15.45 ± 4.54 mm, respectively, but there was no significant correlation between them. We also found no significant differences among gender subgroups for the investigated measures. It is thus important to take the inter-subject variability of skull conductivity and thickness into account by means of using subject-specific calibrated realistic head modeling.
Our goal was to determine the influence of sex, age and the head/brain size on the compartmental brain volumes in the radiologically verified healthy population (96 subjects; 54 women and 42 men) from the Upper Silesia region in Poland. The MRI examinations were done using 3T Philips Achieva with the same T1-weighted and T2-weighted protocols. The image segmentation procedures were performed with SPM (Statistical Parameter Mapping) and FSL-FIRST software. The volumes of 14 subcortical structures for the left and right hemispheres and 4 overall volumes were calculated. The General Linear Models (GLM) analysis was used with and without the Total Brain Volume (TBV) and Intracranial Volume (ICV) parameters as the covariates to study the regional vs. global brain atrophy. After the ICV/TBV adjustments, the majority of sex differences in the specific volumes of interest (VOIs) revealed to be linked to the difference in the head/brain size parameters. The analysis also confirmed the significant effect of the aging process on the brain loss. After the TBV adjustment, the age- and sex-related volumetric trends for the gray and white matter volumes were observed: the negative age dependence of the gray matter volume is more pronounced in the males, while in case of the white matter the positive age-related trend in the female group is weaker. The local losses of the left caudate nucleus and the right thalamus are more advanced than the global brain atrophy. Different head-size correction strategies are not interchangeable and may yield various volumetric results, but when used together, facilitate studies on the regional dependencies inherent to a healthy, but aging, brain.
Age-related brain changes likely contribute to mobility impairments, but the specific mechanisms are poorly understood. Current brain measurement approaches (e.g., functional magnetic resonance imaging (fMRI), functional near infrared spectroscopy (fNIRS), PET) are limited by inability to measure activity from the whole brain during walking. The Mind in Motion Study will use cutting edge, mobile, high-density electroencephalography (EEG). This approach relies upon innovative hardware and software to deliver three-dimensional localization of active cortical and subcortical regions with good spatial and temporal resolution during walking. Our overarching objective is to determine age-related changes in the central neural control of walking and correlate these findings with a comprehensive set of mobility outcomes (clinic-based, complex walking, and community mobility measures). Our hypothesis is that age-related walking deficits are explained in part by the Compensation Related Utilization of Neural Circuits Hypothesis (CRUNCH). CRUNCH is a well-supported model that describes the over-recruitment of brain regions exhibited by older adults in comparison to young adults, even at low levels of task complexity. CRUNCH also describes the limited brain reserve resources available with aging. These factors cause older adults to quickly reach a ceiling in brain resources when performing tasks of increasing complexity, leading to poor performance. Two hundred older adults and twenty young adults will undergo extensive baseline neuroimaging and walking assessments. Older adults will subsequently be followed for up to 3 years. Aim 1 will evaluate whether brain activity during actual walking explains mobility decline. Cross sectional and longitudinal designs will be used to study whether poorer walking performance and steeper trajectories of decline are associated with CRUNCH indices. Aim 2 is to harmonize high-density EEG during walking with fNIRS (during actual and imagined walking) and fMRI (during imagined walking). This will allow integration of CRUNCH-related hallmarks of brain activity across neuroimaging modalities, which is expected to lead to more widespread application of study findings. Aim 3 will study central and peripheral mechanisms (e.g., cerebral blood flow, brain regional volumes, and connectivity, sensory function) to explain differences in CRUNCH indices during walking. Research performed in the Mind in Motion Study will comprehensively characterize the aging brain during walking for developing new intervention targets.
Electroencephalography (EEG) and source estimation can be used to identify brain areas activated during a task, which could offer greater insight on cortical dynamics. Source estimation requires knowledge of the locations of the EEG electrodes. This could be provided with a template or obtained by digitizing the EEG electrode locations. Operator skill and inherent uncertainties of a digitizing system likely produce a range of digitization reliabilities, which could affect source estimation and the interpretation of the estimated source locations. Here, we compared the reliabilities of five digitizing methods (ultrasound, structured-light 3D scan, infrared 3D scan, motion capture probe, and motion capture) and determined the relationship between digitization reliability and source estimation uncertainty, assuming other contributors to source estimation uncertainty were constant. We digitized a mannequin head using each method five times and quantified the reliability and validity of each method. We created five hundred sets of electrode locations based on our reliability results and applied a dipole fitting algorithm (DIPFIT) to perform source estimation. The motion capture method, which recorded the locations of markers placed directly on the electrodes had the best reliability with an average electrode variability of 0.001 cm. Then, in order of decreasing reliability were the method using a digitizing probe in the motion capture system, an infrared 3D scanner, a structured-light 3D scanner, and an ultrasound digitization system. Unsurprisingly, uncertainty of the estimated source locations increased with greater variability of EEG electrode locations and less reliable digitizing systems. If EEG electrode location variability was ∽1 cm, a single source could shift by as much as 2 cm. To help translate these distances into practical terms, we quantified Brodmann area accuracy for each digitizing method and found that the average Brodmann area accuracy for all digitizing methods was >80%. Using a template of electrode locations reduced the Brodmann area accuracy to ∽50%. Overall, more reliable digitizing methods can reduce source estimation uncertainty, but the significance of the source estimation uncertainty depends on the desired spatial resolution. For accurate Brodmann area identification, any of the digitizing methods tested can be used confidently.
Reliable EEG source analysis depends on sufficiently detailed and accurate head models. In this study, we investigate how uncertainties inherent to the experimentally determined conductivity values of the different conductive compartments influence the results of EEG source analysis. In a single source scenario, the superficial and focal somatosensory P20/N20 component, we analyze the influence of varying conductivities on dipole reconstructions using a generalized polynomial chaos (gPC) approach. We find that in particular the conductivity uncertainties for skin and skull have a significant influence on the EEG inverse solution, leading to variations in source localization by several centimeters. The conductivity uncertainties for gray and white matter were found to have little influence on the source localization, but a strong influence on the strength and orientation of the reconstructed source, respectively. As the CSF conductivity is most accurately determined of all conductivities in a realistic head model, CSF conductivity uncertainties had a negligible influence on the source reconstruction. This small uncertainty is a further benefit of distinguishing the CSF in realistic volume conductor models.
Electromagnetic source characterisation requires accurate volume conductor models representing head geometry and the electrical conductivity field. Head tissue conductivity is often assumed from previous literature, however, despite extensive research, measurements are inconsistent. A meta-analysis of reported human head electrical conductivity values was therefore conducted to determine significant variation and subsequent influential factors. Of 3121 identified publications spanning three databases, 56 papers were included in data extraction. Conductivity values were categorised according to tissue type, and recorded alongside methodology, measurement condition, current frequency, tissue temperature, participant pathology and age. We found variation in electrical conductivity of the whole-skull, the spongiform layer of the skull, isotropic, perpendicularly- and parallelly-oriented white matter (WM) and the brain-to-skull-conductivity ratio (BSCR) could be significantly attributed to a combination of differences in methodology and demographics. This large variation should be acknowledged, and care should be taken when creating volume conductor models, ideally constructing them on an individual basis, rather than assuming them from the literature. When personalised models are unavailable, it is suggested weighted average means from the current meta-analysis are used. Assigning conductivity as: 0.41 S/m for the scalp, 0.02 S/m for the whole skull, or when better modelled as a three-layer skull 0.048 S/m for the spongiform layer, 0.007 S/m for the inner compact and 0.005 S/m for the outer compact, as well as 1.71 S/m for the CSF, 0.47 S/m for the grey matter, 0.22 S/m for WM and 50.4 for the BSCR.
In this work, the propagation of electric potentials from source to measurement points was investigated with three finite element method (FEM) models which were based on magnetic resonance imaging (MRI) data sets of increasing spatial resolution. The spatial observation scales of the electrophysiological imaging modalities of electroencephalography (EEG), micro EEG (μEEG), electrocorticography (ECoG) and micro ECoG (μECoG) were modeled. The Leitmotiv of this work is to understand how neuronal and other sources are reflected in high-resolution multi-scale electrophysiological neuroimaging. In the process, anatomical details like foramina, craniotomy defects and cortical depth of modeled dipoles were specifically addressed.
In the first part of this work, it was investigated how electrical potentials propagate from outside of the skull into the brain. This approach is based on the observation that in subdurally implanted epilepsy patients electromyographic (EMG) artifacts are measurable within the skull with ECoG electrodes. To simulate the propagation of those artifacts from the temporal muscle to the ECoG electrodes, we created an individualized FEM model based on MRI data recorded before and after the electrode implantation. Craniotomy defects and highly insulating silicon ECoG grid were implemented. By relating simulation results and measurements in epilepsy patients, it was possible to show that propagation of EMG artifacts into the brain is indeed facilitated by the presence of craniotomy defects, but that even in their absence, the artifacts would still generate 0.11V/m.
In the second part of this work, propagation of cortical potentials to the surface of the head was investigated. To this end, naturally present skull foramina were modeled. Because the spatial resolution of standard 1.5 Tesla (T) diagnostic MRI is not sufficient to resolve skull foramina induced by highly conductive blood vessels which might influence electric propagation, 7T MRI measurements were used to create for the first time a ultra-high resolution FEM model. EEG and μEEG electrodes were added to it. The results of the simulations show that foramina located close to the electrodes have a dual effect. They increase the amplitudes of potentials measured on the surface of the head and distort the electrical field into their direction. The second effect could only be measured using μEEG, the spatial resolution of EEG being to low to capture the electric field shift. This finding highlights the usefulness of spatial multiscale approaches.
After having simulated EEG, μEEG, and ECoG measurements, the last part of this work investigates spatially highly resolved μECoG measurements. As μECoG is not yet medically established, no human measurements exist for comparison. We therefore created a Göttingen minipig brain FEM model. Comparison of in vivo measurements and simulations show that, in Göttingen minipigs, dipolar cortical activity generated at a depth of 0.6mm best fits the spatial profiles measured by μECoG.
In summary the present thesis shows that (i) placement of cortical dipoles at realistic depths and (ii) incorporation of individual skull anatomy, down to millimeter wide foramina, are necessary for realistic multiscale modeling both of electrophysiological signal propagation to and from the brain.
Objective. Our purpose was to evaluate the ability of a dual electrode approach to remove motion artifact from electroencephalography (EEG) measurements. Approach. We used a phantom human head model and robotic motion platform to induce motion while collecting scalp EEG. We assembled a dual electrode array capturing (a) artificial neural signals plus noise from scalp EEG electrodes, and (b) electrically isolated motion artifact noise. We recorded artificial neural signals broadcast from antennae in the phantom head during continuous vertical sinusoidal movements (stationary, 1.00, 1.25, 1.50, 1.75, 2.00 Hz movement frequencies). We evaluated signal quality using signal-to-noise ratio (SNR), cross-correlation, and root mean square error (RMSE) between the ground truth broadcast signals and the recovered EEG signals. Main results. Signal quality was restored following noise cancellation when compared to single electrode EEG measurements collected with no phantom head motion. Significance. We achieved substantial motion artifact attenuation using secondary electrodes for noise cancellation. These methods can be applied to studying electrocortical signals during human locomotion to improve real-world neuroimaging using EEG.
Background:
Accurately solving the electroencephalography (EEG) forward problem is crucial for precise EEG source analysis. Previous studies have shown that the use of multicompartment head models in combination with the finite element method (FEM) can yield high accuracies both numerically and with regard to the geometrical approximation of the human head. However, the workload for the generation of multicompartment head models has often been too high and the use of publicly available FEM implementations too complicated for a wider application of FEM in research studies. In this paper, we present a MATLAB-based pipeline that aims to resolve this lack of easy-to-use integrated software solutions. The presented pipeline allows for the easy application of five-compartment head models with the FEM within the FieldTrip toolbox for EEG source analysis.
Methods:
The FEM from the SimBio toolbox, more specifically the St. Venant approach, was integrated into the FieldTrip toolbox. We give a short sketch of the implementation and its application, and we perform a source localization of somatosensory evoked potentials (SEPs) using this pipeline. We then evaluate the accuracy that can be achieved using the automatically generated five-compartment hexahedral head model [skin, skull, cerebrospinal fluid (CSF), gray matter, white matter] in comparison to a highly accurate tetrahedral head model that was generated on the basis of a semiautomatic segmentation with very careful and time-consuming manual corrections.
Results:
The source analysis of the SEP data correctly localizes the P20 component and achieves a high goodness of fit. The subsequent comparison to the highly detailed tetrahedral head model shows that the automatically generated five-compartment head model performs about as well as a highly detailed four-compartment head model (skin, skull, CSF, brain). This is a significant improvement in comparison to a three-compartment head model, which is frequently used in praxis, since the importance of modeling the CSF compartment has been shown in a variety of studies.
Conclusion:
The presented pipeline facilitates the use of five-compartment head models with the FEM for EEG source analysis. The accuracy with which the EEG forward problem can thereby be solved is increased compared to the commonly used three-compartment head models, and more reliable EEG source reconstruction results can be obtained.
In source localization of electroencephalograpic (EEG) signals, as well as in targeted transcranial current stimulation (TCS), a volume conductor model is required to describe the flow of electric currents in the head. Boundary element models (BEM) can be readily computed to represent major tissue compartments, but cannot encode detailed anatomical information within compartments. Finite element models (FEM) can capture more tissue types and intricate anatomical structures, but with the higher precision also comes the need for semi-automated segmentation, and a higher computational cost. In either case, adjusting to the individual human anatomy requires costly magnetic resonance imaging (MRI), and thus head modeling is often based on the anatomy of an 'arbitrary' individual (e.g. Colin27). Additionally, existing reference models for the human head often do not include the cerebro-spinal fluid (CSF), and their field of view excludes portions of the head and neck - two factors that demonstrably affect current-flow patterns. Here we present a highly detailed FEM, which we call ICBM-NY. It is based on the ICBM152 anatomical template (a non-linear average of the MRI of 152 adult human brains), for which we extended the field of view to the neck, and performed a detailed segmentation of six tissue types (scalp, skull, CSF, gray matter, white matter, air cavities) at 0.5mm(3) resolution. The model was solved for 231 electrode locations. To evaluate its performance, additional FEMs and BEMs were constructed for four individual subjects. Each of the four individual FEMs (regarded as the 'ground truth') is compared to its BEM counterpart, the ICBM-NY, a BEM of the ICBM anatomy, an 'individualized' BEM of the ICBM anatomy warped to the individual head surface, and FEMs of the other individuals. Performance is measured in terms of EEG source localization and TCS targeting errors. Results show that the ICBM-NY outperforms FEMs of mismatched individual anatomies as well as the BEM of the ICBM anatomy according to both criteria. We therefore propose the ICBM-NY as a new standard head model to be used in future EEG and TCS studies whenever an individual MRI is not available. We release all model data online to facilitate broad adoption.
Objective. Transcranial direct current stimulation (tDCS) is a non-invasive brain stimulation technique to modify neural excitability. Using multi-array tDCS, we investigate the influence of inter-individually varying head tissue conductivity profiles on optimal electrode configurations for an auditory cortex stimulation. Approach. In order to quantify the uncertainty of the optimal electrode configurations, multi-variate generalized polynomial chaos expansions of the model solutions are used based on uncertain conductivity profiles of the compartments skin, skull, gray matter, and white matter. Stochastic measures, probability density functions, and sensitivity of the quantities of interest are investigated for each electrode and the current density at the target with the resulting stimulation protocols visualized on the head surface. Main results. We demonstrate that the optimized stimulation protocols are only comprised of a few active electrodes, with tolerable deviations in the stimulation amplitude of the anode. However, large deviations in the order of the uncertainty in the conductivity profiles could be noted in the stimulation protocol of the compensating cathodes. Regarding these main stimulation electrodes, the stimulation protocol was most sensitive to uncertainty in skull conductivity. Finally, the probability that the current density amplitude in the auditory cortex target region is supra-threshold was below 50%. Significance. The results suggest that an uncertain conductivity profile in computational models of tDCS can have a substantial influence on the prediction of optimal stimulation protocols for stimulation of the auditory cortex. The investigations carried out in this study present a possibility to predict the probability of providing a therapeutic effect with an optimized electrode system for future auditory clinical and experimental procedures of tDCS applications.
To increase the reliability for the non-invasive determination of the irritative zone in presurgical epilepsy diagnosis, we introduce here a new experimental and methodological source analysis pipeline that combines the complementary information in EEG and MEG, and apply it to data from a patient, suffering from refractory focal epilepsy. Skull conductivity parameters in a six compartment finite element head model with brain anisotropy, constructed from individual MRI data, are estimated in a calibration procedure using somatosensory evoked potential (SEP) and field (SEF) data. These data are measured in a single run before acquisition of further runs of spontaneous epileptic activity. Our results show that even for single interictal spikes, volume conduction effects dominate over noise and need to be taken into account for accurate source analysis. While cerebrospinal fluid and brain anisotropy influence both modalities, only EEG is sensitive to skull conductivity and conductivity calibration significantly reduces the difference in especially depth localization of both modalities, emphasizing its importance for combining EEG and MEG source analysis. On the other hand, localization differences which are due to the distinct sensitivity profiles of EEG and MEG persist. In case of a moderate error in skull conductivity, combined source analysis results can still profit from the different sensitivity profiles of EEG and MEG to accurately determine location, orientation and strength of the underlying sources. On the other side, significant errors in skull modeling are reflected in EEG reconstruction errors and could reduce the goodness of fit to combined datasets. For combined EEG and MEG source analysis, we therefore recommend calibrating skull conductivity using additionally acquired SEP/SEF data.
Objective. High-definition transcranial direct current stimulation (HD-tDCS) and high-density electroencephalography require accurate models of current flow for precise targeting and current source reconstruction. At a minimum, such modeling must capture the idiosyncratic anatomy of the brain, cerebrospinal fluid (CSF) and skull for each individual subject. Currently, the process to build such high-resolution individualized models from structural magnetic resonance images requires labor-intensive manual segmentation, even when utilizing available automated segmentation tools. Also, accurate placement of many high-density electrodes on an individual scalp is a tedious procedure. The goal was to develop fully automated techniques to reduce the manual effort in such a modeling process. Approach. A fully automated segmentation technique based on Statical Parametric Mapping 8, including an improved tissue probability map and an automated correction routine for segmentation errors, was developed, along with an automated electrode placement tool for high-density arrays. The performance of these automated routines was evaluated against results from manual segmentation on four healthy subjects and seven stroke patients. The criteria include segmentation accuracy, the difference of current flow distributions in resulting HD-tDCS models and the optimized current flow intensities on cortical targets.Main results. The segmentation tool can segment out not just the brain but also provide accurate results for CSF, skull and other soft tissues with a field of view extending to the neck. Compared to manual results, automated segmentation deviates by only 7% and 18% for normal and stroke subjects, respectively. The predicted electric fields in the brain deviate by 12% and 29% respectively, which is well within the variability observed for various modeling choices. Finally, optimized current flow intensities on cortical targets do not differ significantly.Significance. Fully automated individualized modeling may now be feasible for large-sample EEG research studies and tDCS clinical trials.
Subject-specific four-layer boundary element method (BEM) electrical forward head models for four participants, generated from magnetic resonance (MR) head images using NFT ( www.sccn.ucsd.edu/wiki/NFT ), were used to simulate electroencephalographic (EEG) scalp potentials at 256 recorded electrode positions produced by single current dipoles of a 3-D grid in brain space. Locations of these dipoles were then estimated using gradient descent within five template head models fit to the electrode positions. These were: a spherical model, three-layer and four-layer BEM head models based on the Montreal Neurological Institute (MNI) template head image, and these BEM models warped to the recorded electrode positions. Smallest localization errors (4.1-6.2 mm, medians) were obtained using the electrode-position warped four-layer BEM models, with largest localization errors (~20 mm) for most basal brain locations. When we increased the brain-to-skull conductivity ratio assumed in the template model scalp projections from the simulated value (25:1) to a higher value (80:1) used in earlier studies, the estimated dipole locations moved outwards (12.4 mm, median). We also investigated the effects of errors in co-registering the electrode positions, of reducing electrode counts, and of adding a fifth, isotropic white matter layer to one individual head model. Results show that when individual subject MR head images are not available to construct subject-specific head models, accurate EEG source localization should employ a four- or five-layer BEM template head model incorporating an accurate skull conductivity estimate and warped to 64 or more accurately 3-D measured and co-registered electrode positions.
Independent component analysis (ICA) and blind source separation (BSS) methods are increasingly used to separate individual brain and non-brain source signals mixed by volume conduction in electroencephalographic (EEG) and other electrophysiological recordings. We compared results of decomposing thirteen 71-channel human scalp EEG datasets by 22 ICA and BSS algorithms, assessing the pairwise mutual information (PMI) in scalp channel pairs, the remaining PMI in component pairs, the overall mutual information reduction (MIR) effected by each decomposition, and decomposition 'dipolarity' defined as the number of component scalp maps matching the projection of a single equivalent dipole with less than a given residual variance. The least well-performing algorithm was principal component analysis (PCA); best performing were AMICA and other likelihood/mutual information based ICA methods. Though these and other commonly-used decomposition methods returned many similar components, across 18 ICA/BSS algorithms mean dipolarity varied linearly with both MIR and with PMI remaining between the resulting component time courses, a result compatible with an interpretation of many maximally independent EEG components as being volume-conducted projections of partially-synchronous local cortical field activity within single compact cortical domains. To encourage further method comparisons, the data and software used to prepare the results have been made available (http://sccn.ucsd.edu/wiki/BSSComparison).
We derive an asymptotic Newton algorithm for quasi-maximum likelihood estimation of the ICA mixture model, using the ordinary gradient and Hessian. The probabilistic mixture framework yields an algorithm that can accommodate non-stationary environments and arbitrary source densities. We prove asymptotic stability when the source models match the true sources. An example application to EEC segmentation is given.
Bioelectric source measurements are influenced by the measurement location as well as the conductive
properties of the tissues. Volume conductor effects such as the poorly conducting bones or the moderately conducting skin are known to affect the measurement precision and accuracy of the surface electroencephalography (EEG) measurements. This paper investigates the influence of age via skull conductivity upon surface and subdermal bipolar EEG measurement sensitivity conducted on two realistic head models from the Visible Human Project. Subdermal electrodes (a.k.a. subcutaneous
electrodes) are implanted on the skull beneath the skin, fat, and muscles. We studied the effect of age upon these two electrode types according to the scalp-to-skull conductivity ratios of 5, 8, 15, and 30 : 1. The effects on the measurement sensitivity were studied by means of the half-sensitivity volume (HSV) and the region of interest sensitivity ratio (ROISR). The results indicate that the subdermal implantation notably enhances the precision and accuracy of EEG measurements by a factor of eight compared to the scalp surface measurements. In summary, the evidence indicates that both surface and subdermal EEG measurements benefit better recordings in terms of precision and accuracy on younger patients.
In this primer, we give a review of the inverse problem for EEG source localization. This is intended for the researchers new in the field to get insight in the state-of-the-art techniques used to find approximate solutions of the brain sources giving rise to a scalp potential recording. Furthermore, a review of the performance results of the different techniques is provided to compare these different inverse solutions. The authors also include the results of a Monte-Carlo analysis which they performed to compare four non parametric algorithms and hence contribute to what is presently recorded in the literature. An extensive list of references to the work of other researchers is also provided.
This paper starts off with a mathematical description of the inverse problem and proceeds to discuss the two main categories of methods which were developed to solve the EEG inverse problem, mainly the non parametric and parametric methods. The main difference between the two is to whether a fixed number of dipoles is assumed a priori or not. Various techniques falling within these categories are described including minimum norm estimates and their generalizations, LORETA, sLORETA, VARETA, S-MAP, ST-MAP, Backus-Gilbert, LAURA, Shrinking LORETA FOCUSS (SLF), SSLOFO and ALF for non parametric methods and beamforming techniques, BESA, subspace techniques such as MUSIC and methods derived from it, FINES, simulated annealing and computational intelligence algorithms for parametric methods. From a review of the performance of these techniques as documented in the literature, one could conclude that in most cases the LORETA solution gives satisfactory results. In situations involving clusters of dipoles, higher resolution algorithms such as MUSIC or FINES are however preferred. Imposing reliable biophysical and psychological constraints, as done by LAURA has given superior results. The Monte-Carlo analysis performed, comparing WMN, LORETA, sLORETA and SLF, for different noise levels and different simulated source depths has shown that for single source localization, regularized sLORETA gives the best solution in terms of both localization error and ghost sources. Furthermore the computationally intensive solution given by SLF was not found to give any additional benefits under such simulated conditions.
Ventricular enlargement may be an objective and sensitive measure of neuropathological change associated with mild cognitive impairment (MCI) and Alzheimer's disease (AD), suitable to assess disease progression for multi-centre studies. This study compared (i) ventricular enlargement after six months in subjects with MCI, AD and normal elderly controls (NEC) in a multi-centre study, (ii) volumetric and cognitive changes between Apolipoprotein E genotypes, (iii) ventricular enlargement in subjects who progressed from MCI to AD, and (iv) sample sizes for multi-centre MCI and AD studies based on measures of ventricular enlargement. Three dimensional T(1)-weighted MRI and cognitive measures were acquired from 504 subjects (NEC n = 152, MCI n = 247 and AD n = 105) participating in the multi-centre Alzheimer's Disease Neuroimaging Initiative. Cerebral ventricular volume was quantified at baseline and after six months using semi-automated software. For the primary analysis of ventricle and neurocognitive measures, between group differences were evaluated using an analysis of covariance, and repeated measures t-tests were used for within group comparisons. For secondary analyses, all groups were dichotomized for Apolipoprotein E genotype based on the presence of an epsilon 4 polymorphism. In addition, the MCI group was dichotomized into those individuals who progressed to a clinical diagnosis of AD, and those subjects that remained stable with MCI after six months. Group differences on neurocognitive and ventricle measures were evaluated by independent t-tests. General sample size calculations were computed for all groups derived from ventricle measurements and neurocognitive scores. The AD group had greater ventricular enlargement compared to both subjects with MCI (P = 0.0004) and NEC (P < 0.0001), and subjects with MCI had a greater rate of ventricular enlargement compared to NEC (P = 0.0001). MCI subjects that progressed to clinical AD after six months had greater ventricular enlargement than stable MCI subjects (P = 0.0270). Ventricular enlargement was different between Apolipoprotein E genotypes within the AD group (P = 0.010). The number of subjects required to demonstrate a 20% change in ventricular enlargement was substantially lower than that required to demonstrate a 20% change in cognitive scores. Ventricular enlargement represents a feasible short-term marker of disease progression in subjects with MCI and subjects with AD for multi-centre studies.
An examination is made of dipole location errors in electroencephalogram (EEG) source analysis, due to not incorporating the ventricular system (VS), omitting a hole in the skull and underestimating skull conductivity. The simulations are performed for a large number of test dipoles in 3D using the finite difference method. The maximum dipole location error encountered, utilising 27 and 53 electrodes is 7.6 mm and 6.1 mm, respectively when omitting the VS, 5.6 mm and 5.2 mm, respectively when neglecting the hole in the skull, and 33.4 mm and 28.0 mm, respectively when underestimating skull conductivity. The largest location errors due to neglecting the VS can be found in the vicinity of the VS. The largest location errors due to omitting a hole can be found in the vicinity of the hole. At these positions the fitted dipoles are found close to the hole. When skull conductivity is underestimated, the dipole is fitted close to the skull-brain border in a radial direction for all test dipoles. It was found that the location errors due to underestimating skull conductivity are typically higher than those found due to neglecting the VS or neglecting a hole in the skull.
Objective
To determine whether dipoles are an appropriate simplified representation of neural sources for stereo-EEG (sEEG).
Methods
We compared the distributions of voltages generated by a dipole, biophysically realistic cortical neuron models, and extended regions of cortex to determine how well a dipole represented neural sources at different spatial scales and at electrode to neuron distances relevant for sEEG. We also quantified errors introduced by the dipole approximation of neural sources in sEEG source localization using standardized low-resolution electrotomography (sLORETA).
Results
For pyramidal neurons, the coefficient of correlation between voltages generated by a dipole and neuron model were >0.9 for distances >1 mm. For small regions of cortex (∼0.1 cm²), the error in voltages between a dipole and region was <100 µV for all distances. However, larger regions of active cortex (>5 cm²) yielded >50 µV errors within 1.5 cm of an electrode when compared to single dipoles. Finally, source localization errors were <5 mm when using dipoles to represent realistic neural sources.
Conclusions
Single dipoles are an appropriate source model to represent both single neurons and small regions of active cortex, while multiple dipoles are required to represent large regions of cortex.
Significance
Dipoles are computationally tractable and valid source models for sEEG.
Model calibration measures the agreement between the predicted probability estimates and the true correctness likelihood. Proper model calibration is vital for high-risk applications. Unfortunately, modern deep neural networks are poorly calibrated, compromising trustworthiness and reliability. Medical image segmentation particularly suffers from this due to the natural uncertainty of tissue boundaries. This is exasperated by their loss functions, which favor overconfidence in the majority classes. We address these challenges with DOMINO, a domain-aware model calibration method that leverages the semantic confusability and hierarchical similarity between class labels. Our experiments demonstrate that our DOMINO-calibrated deep neural networks outperform non-calibrated models and state-of-the-art morphometric methods in head image segmentation. Our results show that our method can consistently achieve better calibration, higher accuracy, and faster inference times than these methods, especially on rarer classes. This performance is attributed to our domain-aware regularization to inform semantic model calibration. These findings show the importance of semantic ties between class labels in building confidence in deep learning models. The framework has the potential to improve the trustworthiness and reliability of generic medical image segmentation models. The code for this article is available at: https://github.com/lab-smile/DOMINO.
KeywordsImage segmentationMachine learning uncertaintyModel calibrationModel generalizabilityWhole head MRI
Inaccurate estimation of skull conductivity is the largest impediment to high-resolution EEG source imaging because of its strong influence and wide variability across individuals. Nonetheless, there is yet no widely applied method for noninvasively measuring individual skull conductivity. We presented a skull conductivity and source location estimation algorithm (SCALE) for simultaneously estimating skull conductivity and the cortical distributions of 18-20 effective sources derived from the EEG data by independent component analysis (ICA). SCALE combines a realistic Finite Element Method (FEM) head model built from a magnetic resonance (MR) head image with the effective source scalp maps to estimate brain-to-skull conductivity ratio (BSCR) and to map the effective sources on the cortical surface. To estimate the robustness of SCALE BSCR estimates, we applied SCALE to MR image and high-density EEG data from ten participants, five having data from 2-3 different tasks and sessions. As expected, across participants SCALE BSCR estimates differed widely (mean 32.8, range 18-78). Within-participant SCALE BSCR estimates were far more consistent than between participants. By incorporating SCALE-optimized distributed EEG source localization, stable functional imaging of cortical EEG effective sources can become routine, giving relatively low-cost EEG imaging a spatial resolution compatible with other brain imaging results and uniquely capable for studying brain dynamics supporting thought and action in laboratory, virtual, and natural environments.
Given the worldwide increasing socioeconomic burden of aging-associated brain diseases, there is pressing need to gain in-depth knowledge about the neurobiology of brain anatomy changes across the life span. Advances in quantitative magnetic resonance imaging sensitive to brain's myelin, iron, and free water content allow for a detailed in vivo investigation of aging-related changes while reducing spurious morphometry differences. Main aim of our study is to link previous morphometry findings in aging to microstructural tissue properties in a large-scale cohort (n = 966, age range 46-86 y). Addressing previous controversies in the field, we present results obtained with different approaches to adjust local findings for global effects. Beyond the confirmation of age-related atrophy, myelin, and free water decreases, we report proportionally steeper volume, iron, and myelin decline in sensorimotor and subcortical areas paralleled by free water increase. We demonstrate aging-related white matter volume, myelin, and iron loss in frontostriatal projections. Our findings provide robust evidence for spatial overlap between volume and tissue property differences in aging that affect predominantly motor and executive networks.
The electroencephalogram (EEG) provides a non-invasive, minimally restrictive, and relatively low-cost measure of mesoscale brain dynamics with high temporal resolution. Although signals recorded in parallel by multiple, near-adjacent EEG scalp electrode channels are highly-correlated and combine signals from many different sources, biological and non-biological, independent component analysis (ICA) has been shown to isolate the various source generator processes underlying those recordings. Independent components (IC) found by ICA decomposition can be manually inspected, selected, and interpreted, but doing so requires both time and practice as ICs have no order or intrinsic interpretations and therefore require further study of their properties. Alternatively, sufficiently-accurate automated IC classifiers can be used to classify ICs into broad source categories, speeding the analysis of EEG studies with many subjects and enabling the use of ICA decomposition in near-real-time applications. While many such classifiers have been proposed recently, this work presents the ICLabel project comprised of (1) the ICLabel dataset containing spatiotemporal measures for over 200,000 ICs from more than 6000 EEG recordings and matching component labels for over 6000 of those ICs, all using common average reference, (2) the ICLabel website for collecting crowdsourced IC labels and educating EEG researchers and practitioners about IC interpretation, and (3) the automated ICLabel classifier, freely available for MATLAB. The ICLabel classifier improves upon existing methods in two ways: by improving the accuracy of the computed label estimates and by enhancing its computational efficiency. The classifier outperforms or performs comparably to the previous best publicly available automated IC component classification method for all measured IC categories while computing those labels ten times faster than that classifier as shown by a systematic comparison against other publicly available EEG IC classifiers.
The individual alpha peak frequency (IAPF) of the human EEG typically experiences slowing with increasing age. Despite this hallmark change, studies that investigate modulations of conventional EEG alpha power and connectivity by aging and age-related neuropathology neglect to account for inter-group differences in IAPF. To investigate the relationship of age-related IAPF slowing with EEG power and connectivity, we recorded eyes-closed resting state EEG in 37 young adults and 32 older adults. We replicated the finding of a slowed IAPF in older adults. IAPF values were significantly correlated with the frequency of maximum global connectivity and the means of their distributions did not differ, suggesting that connectivity was highest at the IAPF. Older adults expressed reduced global EEG power and connectivity at the conventional upper alpha band (10-12 Hz) compared to young adults. In contrast, groups had equivalent power and connectivity at the IAPF. The results suggest that conventional spectral boundaries may be biased against older adults or any group with a slowed IAPF. We conclude that investigations of alpha activity in aging and age-related neuropathology should be adapted to the IAPF of the individual and that previous findings should be interpreted with caution. EEG in the dominant alpha range may be unsuitable for examining cortico-cortical connectivity due to its subcortical origins.
Anatomically realistic volume conductor models of the human head are important for accurate forward modeling of the electric field during transcranial brain stimulation (TBS), electro- (EEG) and magnetoencephalography (MEG). In particular, the skull compartment exerts a strong influence on the field distribution due to its low conductivity, suggesting the need to represent its geometry accurately. However, automatic skull reconstruction from structural magnetic resonance (MR) images is difficult, as compact bone has a very low signal in magnetic resonance imaging (MRI). Here, we evaluate three methods for skull segmentation, namely FSL BET2, the unified segmentation routine of SPM12 with extended spatial tissue priors, and the skullfinder tool of BrainSuite. To our knowledge, this study is the first to rigorously assess the accuracy of these state-of-the-art tools by comparison with CT-based skull segmentations on a group of ten subjects. We demonstrate several key factors that improve the segmentation quality, including the use of multi-contrast MRI data, the optimization of the MR sequences and the adaptation of the parameters of the segmentation methods. We conclude that FSL and SPM12 achieve better skull segmentations than BrainSuite. The former methods obtain reasonable results for the upper part of the skull when a combination of T1- and T2-weighted images is used as input. The SPM12-based results can be improved slightly further by means of simple morphological operations to fix local defects. In contrast to FSL BET2, the SPM12-based segmentation with extended spatial tissue priors and the BrainSuite-based segmentation provide coarse reconstructions of the vertebrae, enabling the construction of volume conductor models that include the neck. We exemplarily demonstrate that the extended models enable a more accurate estimation of the electric field distribution during transcranial direct current stimulation (tDCS) for montages that involve extraencephalic electrodes. The methods provided by FSL and SPM12 are integrated into pipelines for the automatic generation of realistic head models based on tetrahedral meshes, which are distributed as part of the open-source software package SimNIBS for field calculations for transcranial brain stimulation.
Reconstruction of source neural activity has an increasing importance due to its high time resolution, promoting its application in diagnosis of neurodegenerative and cognitive tasks. To improve the accuracy of reconstruction, we study the influence of anisotropic blood vessels in the EEG-based source localization solution, including several tissues. To this end, we develop a model that reflects physical properties of the head volume based on collected angiographic data. From obtained results of real data, we find that omission of the anisotropic blood vessels within the forward modeling may result in potential discrepancies larger than 35 \upmu V and dipole localization errors greater than 15 mm, especially, in deep brain areas.
Accurate electroencephalographic (EEG) source localization requires an electrical head model incorporating accurate geometries and conductivity values for the major head tissues. While consistent conductivity values have been reported for scalp, brain, and cerebrospinal fluid, measured brain-to-skull conductivity ratio (BSCR) estimates have varied between 8 and 80, likely reflecting both inter-subject and measurement method differences. In simulations, mis-estimation of skull conductivity can produce source localization errors as large as 3 cm. Here, we describe an iterative gradient-based approach to Simultaneous tissue Conductivity And source Location Estimation (SCALE). The scalp projection maps used by SCALE are obtained from near-dipolar effective EEG sources found by adequate independent component analysis (ICA) decomposition of sufficient high-density EEG data. We applied SCALE to simulated scalp projections of 15 cm(2)-scale cortical patch sources in an MR image-based electrical head model with simulated BSCR of 30. Initialized either with a BSCR of 80 or 20, SCALE estimated BSCR as 32.6. In Adaptive Mixture ICA (AMICA) decompositions of (45-min, 128-channel) EEG data from two young adults we identified sets of 13 independent components having near-dipolar scalp maps compatible with a single cortical source patch. Again initialized with either BSCR 80 or 25, SCALE gave BSCR estimates of 34 and 54 for the two subjects respectively. The ability to accurately estimate skull conductivity non-invasively from any well-recorded EEG data in combination with a stable and non-invasively acquired MR imaging-derived electrical head model could remove a critical barrier to using EEG as a sub-cm(2)-scale accurate 3-D functional cortical imaging modality.
Copyright © 2015. Published by Elsevier Inc.
BACKGROUND: Imaging and postmortem studies provide converging evidence that, beginning in adolescence, gray matter volume declines linearly until old age, while cerebrospinal fluid volumes are stable in adulthood (age 20-50 years). Given the fixed volume of the cranium in adulthood, it is surprising that most studies observe no white matter volume expansion after approximately age 20 years. We examined the effects of the aging process on the frontal and temporal lobes. METHODS: Seventy healthy adult men aged 19 to 76 years underwent magnetic resonance imaging. Coronal images focused on the frontal and temporal lobes were acquired using pulse sequences that maximized gray vs white matter contrast. The volumes of total frontal and temporal lobes as well as the gray and white matter subcomponents were evaluated. RESULTS: Age-related linear loss in gray matter volume in both frontal (r = -0.62, P<.001) and temporal (r = -0.48, P<.001) lobes was confirmed. However, the quadratic function best represented the relationship between age and white matter volume in the frontal (P<.001) and temporal (P<.001) lobes. Secondary analyses indicated that white matter volume increased until age 44 years for the frontal lobes and age 47 years for the temporal lobes and then declined. CONCLUSIONS: The changes in white matter suggest that the adult brain is in a constant state of change roughly defined as periods of maturation continuing into the fifth decade of life followed by degeneration. Pathological states that interfere with such maturational processes could result in neurodevelopmental arrests in adulthood.
For accurate EEG/MEG source analysis it is necessary to model the head volume conductor as realistic as possible. This includes the distinction of the different conductive compartments in the human head. In this study, we investigated the influence of modeling/not modeling the conductive compartments skull spongiosa, skull compacta, cerebrospinal fluid (CSF), gray matter, and white matter and of the inclusion of white matter anisotropy on the EEG/MEG forward solution. Therefore, we created a highly realistic 6-compartment head model with white matter anisotropy and used a state-of-the-art finite element approach. Starting from a 3-compartment scenario (skin, skull, and brain), we subsequently refined our head model by distinguishing one further of the above-mentioned compartments. For each of the generated five head models, we measured the effect on the signal topography and signal magnitude both in relation to a highly resolved reference model and to the model generated in the previous refinement step. We evaluated the results of these simulations using a variety of visualization methods, allowing us to gain a general overview of effect strength, of the most important source parameters triggering these effects, and of the most affected brain regions. Thereby, starting from the 3-compartment approach, we identified the most important additional refinement steps in head volume conductor modeling. We were able to show that the inclusion of the highly conductive CSF compartment, whose conductivity value is well known, has the strongest influence on both signal topography and magnitude in both modalities. We found the effect of gray/white matter distinction to be nearly as big as that of the CSF inclusion, and for both of these steps we identified a clear pattern in the spatial distribution of effects. In comparison to these two steps, the introduction of white matter anisotropy led to a clearly weaker, but still strong, effect. Finally, the distinction between skull spongiosa and compacta caused the weakest effects in both modalities when using an optimized conductivity value for the homogenized compartment. We conclude that it is highly recommendable to include the CSF and distinguish between gray and white matter in head volume conductor modeling. Especially for the MEG, the modeling of skull spongiosa and compacta might be neglected due to the weak effects; the simplification of not modeling white matter anisotropy is admissible considering the complexity and current limitations of the underlying modeling approach.
It is known that incorporating cerebrospinal fluid (CSF) into realistic volume conductor models adds precision for source analysis. However, modeling interior CSF compartments like ventricles or deep sulci creates a complex source space with many deep cavities. Such a fragmented source space can cause problems for dipole fitting and other inverse methods. A solution could be to use a simplified head model, where only the superficial CSF layer between the brain surface and the inner skull boundary is included, while interior CSF compartments are ignored. The present paper aims at investigating if simplified CSF models are sufficiently accurate for forward and inverse solutions. A simulation study using realistic volume conductor models was performed. First, a detailed and anatomically plausible reference model was created. Then, two test models were derived. Test model A ignored CSF completely, while test model B only ignored CSF in deep sulci and the ventricles. Forward computation errors were assessed by directly comparing the potentials computed in the reference and test models. Inverse errors were investigated by performing source reconstruction of single sources in the test models to reconstruct the potentials simulated in the reference model. Large topography (relative difference measure (RDM) > 0.1) and localization errors (> 4mm) were found for superficial sources and sources close to internal CSF compartments for model A. In model B, RDM errors > 0.1 were found for only few sources close to the ventricles and close to sulci of larger extent. Localization errors in model B were below 2 mm for nearly all reference sources. The results suggest that ignoring CSF when creating a head model leads to substantial localization errors. Ignoring only internal CSFfilled compartments, while modeling superficial CSF layers allows source localization with relatively high precision. Thus, avoiding a fragmented source space by not modeling internal CSF compartments is acceptable.
The low-conducting human skull is known to have an especially large influence on electroencephalography (EEG) source analysis. Because of difficulties segmenting the complex skull geometry out of magnetic resonance images, volume conductor models for EEG source analysis might contain inaccuracies and simplifications regarding the geometry of the skull. The computer simulation study presented here investigated the influences of a variety of skull geometry deficiencies on EEG forward simulations and source reconstruction from EEG data. Reference EEG data was simulated in a detailed and anatomically plausible reference model. Test models were derived from the reference model representing a variety of skull geometry inaccuracies and simplifications. These included erroneous skull holes, local errors in skull thickness, modeling cavities as bone, downward extension of the model and simplifying the inferior skull or the inferior skull and scalp as layers of constant thickness. The reference EEG data was compared to forward simulations in the test models, and source reconstruction in the test models was performed on the simulated reference data. The finite element method with high-resolution meshes was employed for all forward simulations. It was found that large skull geometry inaccuracies close to the source space, for example, when cutting the model directly below the skull, led to errors of 20mm and more for extended source space regions. Local defects, for example, erroneous skull holes, caused non-negligible errors only in the vicinity of the defect. The study design allowed a comparison of influence size, and guidelines for modeling the skull geometry were concluded.
There is consistent evidence that brain volume changes in early and late life. Most longitudinal studies usually only span a few years and include a limited number of participants. In this review, we integrate findings from 56 longitudinal magnetic resonance imaging (MRI) studies on whole brain volume change in healthy individuals. The individual longitudinal MRI studies describe only the development in a limited age range. In total, 2,211 participants were included. Age at first measurement varied between 4 and 88 years of age. The studies included in this review were performed using a large range of methods (e.g., different scanner protocols and different acquisition parameters). We applied a weighted regression analysis to estimate the age dependency of the rate of relative annual brain volume change across studies. The results indicate that whole brain volume changes throughout the life span. A wave of growth occurs during childhood/adolescence, where around 9 years of age a 1% annual brain growth is found which levels off until at age 13 a gradual volume decrease sets in. During young adulthood, between ∼18 and 35 years of age, possibly another wave of growth occurs or at least a period of no brain tissue loss. After age 35 years, a steady volume loss is found of 0.2% per year, which accelerates gradually to an annual brain volume loss of 0.5% at age 60. The brains of people over 60 years of age show a steady volume loss of more than 0.5%. Understanding the mechanisms underlying these plastic brain changes may contribute to distinguishing progressive brain changes in psychiatric and neurological diseases from healthy aging processes. Hum Brain Mapp, 2012. © 2011 Wiley Periodicals, Inc.
Although numerous published reports have demonstrated the beneficial effects of transcranial direct-current stimulation (tDCS) on task performance, fundamental questions remain regarding the optimal electrode configuration on the scalp. Moreover, it is expected that lesioned brain tissue will influence current flow and should therefore be considered (and perhaps leveraged) in the design of individualized tDCS therapies for stroke. The current report demonstrates how different electrode configurations influence the flow of electrical current through brain tissue in a patient who responded positively to a tDCS treatment targeting aphasia. The patient, a 60-year-old man, sustained a left hemisphere ischemic stroke (lesion size = 87.42 mL) 64 months before his participation. In this study, we present results from the first high-resolution (1 mm(3)) model of tDCS in a brain with considerable stroke-related damage; the model was individualized for the patient who received anodal tDCS to his left frontal cortex with the reference cathode electrode placed on his right shoulder. We modeled the resulting brain current flow and also considered three additional reference electrode positions: right mastoid, right orbitofrontal cortex, and a "mirror" configuration with the anode over the undamaged right cortex. Our results demonstrate the profound effect of lesioned tissue on resulting current flow and the ability to modulate current pattern through the brain, including perilesional regions, through electrode montage design. The complexity of brain current flow modulation by detailed normal and pathologic anatomy suggest: (1) That computational models are critical for the rational interpretation and design of individualized tDCS stroke-therapy; and (2) These models must accurately reproduce head anatomy as shown here.
The structure of the brain is constantly changing from birth throughout the lifetime, meaning that normal aging, free from dementia, is associated with structural brain changes. This paper reviews recent evidence from magnetic resonance imaging (MRI) studies about age-related changes in the brain. The main conclusions are that (1) the brain shrinks in volume and the ventricular system expands in healthy aging. However, the pattern of changes is highly heterogeneous, with the largest changes seen in the frontal and temporal cortex, and in the putamen, thalamus, and accumbens. With modern approaches to analysis of MRI data, changes in cortical thickness and subcortical volume can be tracked over periods as short as one year, with annual reductions of between 0.5% and 1.0% in most brain areas. (2) The volumetric brain reductions in healthy aging are likely only to a minor extent related to neuronal loss. Rather, shrinkage of neurons, reductions of synaptic spines, and lower numbers of synapses probably account for the reductions in grey matter. In addition, the length of myelinated axons is greatly reduced, up to almost 50%. (3) Reductions in specific cognitive abilities--for instance processing speed, executive functions, and episodic memory--are seen in healthy aging. Such reductions are to a substantial degree mediated by neuroanatomical changes, meaning that between 25% and 100% of the differences between young and old participants in selected cognitive functions can be explained by group differences in structural brain characteristics.
We used computer simulations to investigate finite element models of the layered structure of the human skull in EEG source analysis. Local models, where each skull location was modeled differently, and global models, where the skull was assumed to be homogeneous, were compared to a reference model, in which spongy and compact bone were explicitly accounted for. In both cases, isotropic and anisotropic conductivity assumptions were taken into account. We considered sources in the entire brain and determined errors both in the forward calculation and the reconstructed dipole position. Our results show that accounting for the local variations over the skull surface is important, whereas assuming isotropic or anisotropic skull conductivity has little influence. Moreover, we showed that, if using an isotropic and homogeneous skull model, the ratio between skin/brain and skull conductivities should be considerably lower than the commonly used 80:1. For skull modeling, we recommend (1) Local models: if compact and spongy bone can be identified with sufficient accuracy (e.g., from MRI) and their conductivities can be assumed to be known (e.g., from measurements), one should model these explicitly by assigning each voxel to one of the two conductivities, (2) Global models: if the conditions of (1) are not met, one should model the skull as either homogeneous and isotropic, but with considerably higher skull conductivity than the usual 0.0042 S/m, or as homogeneous and anisotropic, but with higher radial skull conductivity than the usual 0.0042 S/m and a considerably lower radial:tangential conductivity anisotropy than the usual 1:10.
The paper describes finite element related procedures for inverse localization of multiple sources in realistically shaped head models. Dipole sources are modeled by placing proper monopole sources on neighboring nodes. Lead field operators are established for dipole sources. Two different strategies for the solution of inverse problems, namely combinatorial optimization techniques and regularization methods are discussed and applied to visually evoked potentials, for which exemplary results are shown. Most of the procedures described are fully automatic and require only proper input preparation. The overall work for the example presented (from EEG recording to visual inspection of the results) can be performed in roughly a week, most of which is waiting time for the computation of the lead field matrix or inverse calculations on a standard and affordable engineering workstation.
The estimation of cortical current activity from scalp-recorded potentials is a complicated mathematical problem that requires fairly precise knowledge of the location of the scalp electrodes. It is expected that spatial mislocalization of electrodes will introduce errors in this estimation. The present study uses simulated and real data to quantify these errors for dipole current sources in a spherical head model.
A 3-dimensional digitizer was used to locate the positions of 31 scalp electrodes placed on the head according to the 10-20 system in 10 normal subjects. Dipole localizations were performed on auditory evoked potentials (AEPs) collected from these subjects.
Computer simulations with several dipole source configurations suggest that errors in locations and orientations on the order of 5 mm and 5 degrees, respectively, are possible for electrode mislocalizations of about 5 degrees. In actual experimental settings, digitized electrode positions were typically mislocalized by an average of about 4 degrees from their standard 10-20 positions on a spherical model. These differences in electrode positions translated to mean differences of about 8 mm in dipole locations and 5 degrees in dipole orientations.
Dipole estimation errors due to electrode mislocalizations are within the limits of errors due to other modeling approximations and noise.
The conductivity of the human skull was measured both in vitro and in vivo. The in vitro measurement was performed on a sample of fresh skull placed within a saline environment. For the in vivo measurement a small current was passed through the head by means of two electrodes placed on the scalp. The potential distribution thus generated on the scalp was measured in two subjects for two locations of the current injecting electrodes. Both methods revealed a skull conductivity of about 0.015 [symbol: see text]/m. For the conductivities of the brain, the skull and the scalp a ratio of 1:1/15:1 was found. This is consistent with some of the reports on conductivities found in the literature, but differs considerably from the ratio 1:1/80:1 commonly used in neural source localization. An explanation is provided for this discrepancy, indicating that the correct ratio is 1:1/15:1.