Monte Carlo simulation studies of EEG and MEG localization accuracy

Massachusetts General Hospital, NMR Center, Building 149, 13th Street, Charlestown, MA 02129, USA.
Human Brain Mapping (Impact Factor: 5.97). 05/2002; 16(1):47-62. DOI: 10.1002/hbm.10024
Source: PubMed


Both electroencephalography (EEG) and magnetoencephalography (MEG) are currently used to localize brain activity. The accuracy of source localization depends on numerous factors, including the specific inverse approach and source model, fundamental differences in EEG and MEG data, and the accuracy of the volume conductor model of the head (i.e., the forward model). Using Monte Carlo simulations, this study removes the effect of forward model errors and theoretically compares the use of EEG alone, MEG alone, and combined EEG/MEG data sets for source localization. Here, we use a linear estimation inverse approach with a distributed source model and a realistic forward head model. We evaluated its accuracy using the crosstalk and point spread metrics. The crosstalk metric for a specified location on the cortex describes the amount of activity incorrectly localized onto that location from other locations. The point spread metric provides the complementary measure: for that same location, the point spread describes the mis-localization of activity from that specified location to other locations in the brain. We also propose and examine the utility of a "noise sensitivity normalized" inverse operator. Given our particular forward and inverse models, our results show that 1) surprisingly, EEG localization is more accurate than MEG localization for the same number of sensors averaged over many source locations and orientations; 2) as expected, combining EEG with MEG produces the best accuracy for the same total number of sensors; 3) the noise sensitivity normalized inverse operator improves the spatial resolution relative to the standard linear estimation operator; and 4) use of an a priori fMRI constraint universally reduces both crosstalk and point spread.

Download full-text


Available from: Anders M Dale, Sep 30, 2015
15 Reads
  • Source
    • "A recent successful approach to the problem of noise reduction of EEG signals is independent component analysis (ICA), which decomposes a multi-channel signal in a set of sources with maximally independent components (ICs). However, it has limitation on the number of separable ICs, N ICs from N electrodes, which makes the decomposition imperfect as the number of EEG sources is much higher than the number of ICs [3], [4], [5]. In addition, ICA-based methods require subjective decision making [6] or arbitrary tuning of the thresholds [7] to distinguish artifacted ICs from non-artifacted ICs, which is known as the permutation problem. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Data contamination by ocular artifacts such as eye blinks and eye movements is a major barrier that must be overcome when attempting to analyze electroencephalogram (EEG) and event-related potential (ERP) data. To handle this problem, a number of artifact removal methods has been proposed. Specifically, we focus on a method using a multi-channel Wiener filters based on a probabilistic generative model. This method assumes that the observed signal is the sum of multiple signals elicited by psychological or physical events, and separates the observed signal into each event signal using estimated model parameters. Based on this scheme, we have proposed a model parameter estimation method using prior information of each event signal. In this paper, we examine the potential of this model to deal with highly contaminated signals by collecting EEG data intentionally contaminated by eye blinks and relatively clean ERP data, and using them as prior information of each event signal. We conducted an experimental evaluation using a classical attention task. The results showed the proposed method effectively enhances the target ERP component while reducing the contamination caused by eye blinks.
    Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Milan, Italy; 08/2015
  • Source
    • "Thus the performance of any linear estimator and measurement modality can be characterized with the PSFs and CTFs that are independent of the actual measured signals; the response to any source distribution represented by the source space can be expressed as linear combination of PSFs. As the PSF and CTF for each elementary source are identical for the MN estimator used here (Liu et al 2002), we present the results for the PSFs only; the results apply for CTFs as well. PSF is a general description of an imaging system. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Solving the inverse problem of electrocardiography (ECG) and magnetocardiography (MCG) is often referred to as cardiac source imaging. Spatial properties of ECG and MCG as imaging systems are, however, not well known. In this modelling study, we investigate the sensitivity and point-spread function (PSF) of ECG, MCG, and combined ECG+MCG as a function of source position and orientation, globally around the ventricles: signal topographies are modelled using a realistically-shaped volume conductor model, and the inverse problem is solved using a distributed source model and linear source estimation with minimal use of prior information. The results show that the sensitivity depends not only on the modality but also on the location and orientation of the source and that the sensitivity distribution is clearly reflected in the PSF. MCG can better characterize tangential anterior sources (with respect to the heart surface), while ECG excels with normally-oriented and posterior sources. Compared to either modality used alone, the sensitivity of combined ECG+MCG is less dependent on source orientation per source location, leading to better source estimates. Thus, for maximal sensitivity and optimal source estimation, the electric and magnetic measurements should be combined.
    Physics in Medicine and Biology 11/2014; 59(23):7141-7158. DOI:10.1088/0031-9155/59/23/7141 · 2.76 Impact Factor
  • Source
    • "Even though simultaneously recorded MEG and EEG reflect the same primary currents (Hämäläinen et al., 1993), they differ in biophysical properties including large differences in leadfields, signal cancellation, propagation, sensitivity to source orientation and distance to sensors (Ahlfors et al., 2010a; Cuffin, 1990; Irimia et al., 2012; Liu et al., 2002; Marinkovic et al., 2004a). As a consequence, the MEG and EEG reveal different aspects of the underlying neural generators especially when these generators are distributed (Dehghani et al., 2010a, 2010b). "
    [Show abstract] [Hide abstract]
    ABSTRACT: This study examined neurofunctional correlates of reading by modulating semantic, lexical, and orthographic attributes of letter strings. It compared the spatio-temporal activity patterns elicited by real words (RW), pseudowords, orthographically regular, pronounceable nonwords (PN) that carry no meaning, and orthographically illegal, nonpronounceable nonwords (NN). A double-duty lexical decision paradigm instructed participants to detect RW while ignoring nonwords and to additionally respond to words that refer to animals (AW). Healthy social drinkers (N=22) participated in both alcohol (0.6 g/kg ethanol for men, 0.55 g/kg for women) and placebo conditions in a counterbalanced design. Whole-head MEG signals were analyzed with an anatomically-constrained MEG method. Simultaneously acquired ERPs confirm previous evidence. Spatio-temporal MEG estimates to RW and PN are consistent with the highly replicable left-lateralized ventral visual processing stream. However, the PN elicit weaker activity than other stimuli starting at ∼230 ms and extending to the M400 (magnetic equivalent of N400) in the left lateral temporal area, indicating their reduced access to lexicosemantic stores. In contrast, the NN uniquely engage the right hemisphere during the M400. Increased demands on lexicosemantic access imposed by AW result in greater activity in the left temporal cortex starting at ∼230 ms and persisting through the M400 and response preparation stages. Alcohol intoxication strongly attenuates early visual responses occipito-temporally overall. Subsequently, alcohol selectively affects the left prefrontal cortex as a function of orthographic and semantic dimensions, suggesting that it modulates the dynamics of the lexicosemantic processing in a top-down manner, by increasing difficulty of semantic retrieval.
    Brain research 04/2014; 1558. DOI:10.1016/j.brainres.2014.02.030 · 2.84 Impact Factor
Show more