Elsevier

Clinical Neurophysiology

Volume 125, Issue 8, August 2014, Pages 1653-1662
Clinical Neurophysiology

Magnetoencephalography signals are influenced by skull defects

https://doi.org/10.1016/j.clinph.2013.12.099Get rights and content

Highlights

  • We present the first in vivo experimental evidence of the substantial influence of skull defects on MEG.

  • The MEG signal amplitude deviates more if the source is central under the skull defect, whereas the EEG signal amplitude deviates more if the source is under the edge of the defect.

  • Dense spatial sampling reveals high spatial frequencies in MEG and EEG signals due to skull defects that are not detectable with current human helmet-type MEG devices and standard EEG setups.

Abstract

Objective

Magnetoencephalography (MEG) signals had previously been hypothesized to have negligible sensitivity to skull defects. The objective is to experimentally investigate the influence of conducting skull defects on MEG and EEG signals.

Methods

A miniaturized electric dipole was implanted in vivo into rabbit brains. Simultaneous recording using 64-channel EEG and 16-channel MEG was conducted, first above the intact skull and then above a skull defect. Skull defects were filled with agar gels, which had been formulated to have tissue-like homogeneous conductivities. The dipole was moved beneath the skull defects, and measurements were taken at regularly spaced points.

Results

The EEG signal amplitude increased 2–10 times, whereas the MEG signal amplitude reduced by as much as 20%. The EEG signal amplitude deviated more when the source was under the edge of the defect, whereas the MEG signal amplitude deviated more when the source was central under the defect. The change in MEG field-map topography (relative difference measure, RDM = 0.15) was geometrically related to the skull defect edge.

Conclusions

MEG and EEG signals can be substantially affected by skull defects.

Significance

MEG source modeling requires realistic volume conductor head models that incorporate skull defects.

Introduction

The signals acquired by electroencephalography (EEG) and magnetoencephalography (MEG) are due to the electric currents generated by brain activity. The volume current through the tissues inside the head modifies the sensor signals at the head surface. The flow of electric current through the tissues influences the potentials observed at the electrodes in EEG. To a lesser degree, the volume current also influences the magnetic flux observed in MEG. The skull has the most resistive tissue of the head, and therefore has the strongest influence on these techniques. The neurological term breach rhythm describes an increase in the amplitude of alpha, beta, and mu rhythms of the brain that occurs proximal to or over post-surgical skull defects (Cobb et al., 1979). At these locations, the absence of skull tissue allows volume currents to reach the electrodes largely unfiltered and unattenuated. The sharper features of breach activity are easily mistaken for interictal epileptic discharges (Brigo et al., 2011). MEG is reported to be less sensitive than EEG to skull defects (Lee et al., 2010).

Theoretically, EEG and MEG signals should both be distorted by the skull and any defects in it. The degree of distortion depends on the depth, orientation, and extent of the source, as well as the geometry and composition of the skull and the skull defect (Meijs et al., 1987, Hämäläinen and Sarvas, 1989). However, the influence of the skull is estimated to be much less on components of MEG signals than on those of EEG signals, based on considerations under simplified conditions of a flat, layered, inhomogeneous volume conductor (Cohen and Hosaka, 1976) and using a concentric, multilayered, spherical, homogeneous volume conductor (Geselowitz, 1970, Grynzpan and Geselowitz, 1973).

A small number of in vivo animal and post-mortem phantom experiments have examined the influence of skull defects on MEG and EEG signals. Barth et al. (1986) used a physical coaxial dipole to simulate intracerebral currents in a formalin-fixed human cranial specimen that had been filled with conducting jelly. Their visual inspection found the MEG signals above a craniotomy to be qualitatively indistinguishable from those above intact skull. Okada et al., (1999b) conducted EEG and MEG measurements associated with somatic evoked responses in anaesthetized piglets, first over intact skull and then over the dura after a large section of skull was removed (skull on versus skull off). They found no significant difference in MEG signal amplitude or morphology, except for an attenuation of the MEG signal when the skull was removed, which was stronger for deeper sources (25% for a 14 mm-deep source).

The limitations of existing experiments are that (1) the skull defect was filled with non-conducting air (except in (Flemming et al., 2005) for EEG); (2) the skull defect was large compared with the sensor planes (skull-on versus skull-off); and (3) that evoked responses were used, which have a high variability with regard to source position, extent, orientation, and amplitude. Therefore, the objective of this study is to experimentally investigate the influence of conducting skull defects on EEG and MEG signals above and around a skull defect, using a well-defined current source under the middle and edge of the defect and next to it, in an in vivo rabbit model.

Section snippets

Electroencephalography

We constructed a miniaturized 64-channel EEG array (Fig. 1A) with electrodes 1.4 mm apart to achieve sufficient spatial sampling density. Ag/AgCl ball electrodes of 0.6 mm diameter were arranged in a regular grid embedded in silicone. The EEG signals were amplified using two synchronized SynAmps (Compumedics NeuroScan, Charlotte, NC, USA) amplifiers. A sampling rate of 1 kHz was used with an analog 0.3–300 Hz band-pass filter and a 50 Hz notch filter. The EEG array position within a stereotactic

Setup geometries

Successful recordings were made in five animals: four in vivo with a source oriented tangentially to the skull surface, and one post mortem with a radially oriented source. The skull thickness around defect 1 was comparatively homogeneous (Table 2). The CT images confirmed that the three layers of the skull around the defects were differentiable and fully developed. The skull defects were of similar sizes. Among the animals, sources were located at different depths from the inner skull surface

Amplitude

A conducting skull defect has much less influence on flux density map than on potential map; however, the influence is substantial, with a field-map amplitude reduction of approximately 20% compared with intact skull (Fig. 7C). If we consider the deviation at a particular point in the sensor plane, such as the peak amplitude in the difference map (Fig. 8, shift pos. 16, peak negative amplitude −5 127 fT), then this deviation constitutes a −49.2% change in the intact skull MEG signal amplitude at

Conclusion

For both MEG and EEG, we observed substantial signal amplitude and topography changes due to a skull defect. The changes were dependent on the exact geometry of the skull defect and the relative orientation and position of the source. The conductivity of the skull defect had a modulating influence on the local amplitude.

We conclude that a realistic representation of the skull and skull defects in volume conductor models of the head is important for the forward simulation of not only EEG but

Conflict of interest disclosure

All funding sources supporting this work are acknowledged. The authors will disclose to the editor any pertinent financial interests associated with the manufacture of any drug or product described in this manuscript.

Acknowledgments

This work was supported by the German Research Foundation [Ha2899/14-1]; the Australian National Health and Medical Research Council [558425]; the German Academic Exchange Service [D/08/13928, 54388947]; and the Australian Group of Eight. We wish to thank Stefan Clauss, Hannes Nowak, Ralph Huonker, Frank Gießler, Daniel Güllmar, Eric Lopatta, Simon Vogrin, Levin Kuhlmann, David Grayden, and Mark Cook for their support. We thank the Research Workshop of the Jena University Hospital for the

References (27)

  • D.A. Brody

    A theoretical analysis of intracavitary blood mass influence on the heart-lead relationship

    Circ Res

    (1956)
  • L.A. Geddes et al.

    The specific resistance of biological material – A compendium of data for the biomedical engineer and physiologist

    Med Biol Eng Comput

    (1967)
  • D.B. Geselowitz

    On the magnetic field generated outside an inhomogeneous volume conductor by internal current sources

    IEEE Trans Mag

    (1970)
  • Cited by (15)

    • Influence of unfused cranial bones on magnetoencephalography signals in human infants

      2021, Clinical Neurophysiology
      Citation Excerpt :

      However, the conductivity of the fontanels and sutures are higher than the skull conductivity. In this case the skull defects can alter MEG signals more strongly than when they are filled with a poorly conducting material (van den Broek et al., 1998; Lew et al., 2013; Lau et al., 2014, 2016). Such a skull opening can alter the spatial pattern of the MEG signals, the effect depending on the relative location of the source and the edge of the defect.

    • Electrical source imaging of interictal spikes using multiple sparse volumetric priors for presurgical epileptogenic focus localization

      2016, NeuroImage: Clinical
      Citation Excerpt :

      These kind of datasets are however restricted for validation to the locations where the intracranial electrodes are placed. Moreover, simultaneous recordings will affect the ESI results due to skull defects and the placement of the electrodes (Li et al., 2007; Lanfer et al., 2012, 2013; Lau et al., 2014). An alternative way is to evaluate presurgical EEG data with a high incidence of interictal spikes that were recorded in patients with good surgical outcome and who showed no interictal spike activity in postsurgical EEG registrations.

    • Detectability of the somatosensory evoked high frequency oscillation (HFO) co-recorded by scalp EEG and ECoG under propofol

      2016, NeuroImage: Clinical
      Citation Excerpt :

      This is in agreement with computational models (Telenczuk et al., 2015) and experimental evidence (Shimazu et al., 2000), which have shown a dampening factor from epidural to scalp level in the range of 10 to 100 in monkeys. In our surgical setting, skull opening could in principle increase the scalp EEG amplitude (Lau et al., 2014; von Ellenrieder et al., 2014a), which was however not observed in our patient group. We may safely assume that the sources of N20 and HFO are strictly co-localized (Ozaki and Hashimoto, 2011).

    • EEG/MEG Source Reconstruction: Textbook for Electro-and Magnetoencephalography

      2022, EEG/MEG Source Reconstruction: Textbook for Electro-and Magnetoencephalography
    View all citing articles on Scopus
    View full text