Magnetoencephalography: Clinical Uses and Limitations

  1. 1.     Introduction

The average adult human brain (1,350 g) contains approximately 86.1 (±8.1) billion neurons [Azevedo, Carvalho, Grinberg, Farfel, Ferretti, Leite, Filho, Lent & Herculano-Houzel, 2009; Pakkenberg, Pelvig, Marner, Bundgaard, Gundersen, Nyengaard & Regeur, 2003], and the number of non-neuronal glial cells has been speculated to be anywhere from 10 to 50 times the number of neurons [Nishiyama, Yang & Butt, 2005].  Incidentally, some recent comparative research suggests that the ratio of neurons to non-neuronal glial cells may be closer to 1:1 in many areas, and may also fluctuate as a function of location [see Azevedo et al., 2009].  Each of the 86.1 (±8.1) billion neurons in the human brain contains between 103 (1,000) and 104 (10,000) synapses¾depending on the neurons type, location, and function¾which serve to allow it to communicate to other neurons  [for review see Chu-Shore, Kramer, Bianchi, Caviness & Cash, 2011; Drachman, 2005].  It has been estimated that the brain of a 3-year-old child contains about 1015 (1 quadrillion) synapses [Huttenlocher & Dabholkar, 1997], and that the average adult human brain contains between 5 x 1014 (500 trillion) and 7.5 x 1014 (750 trillion) synapses.  Additionally, it has been estimated that the refractory period of a single neuron is between 3-4 milliseconds (ms).  Given the 3-4ms refractory period, and the 7.5 x 1014 synapses, the human brain could potentially perform as many as 2.45 x 1016 (24.5 quadrillion) operations per second.  The sequence and rate of which the synapses between the neurons in our brain are activated are hypothesized to be the way in which we interpret the world, retrieve memories, experience emotions, and potentially the way in which we store memories.  The synapses between the neurons in our brain are also the medium by which neurons are able to send and receive “messages” in the form of neurotransmitters¾chemicals which are released in response to the electrically based depolarization of a neuron, and contribute to either the excitation or inhibition of the neighboring neurons with which it makes synaptic contacts. When neurons experience electrical depolarization the magnetic changes resulting from the electrical oscillations, which reflect neuronal activity, can be measured.

1.1   Source Generation in EEG and MEG

Electroencephalography (EEG) and magnetoencephalography (MEG) are techniques used to measure neuronal activity using sensors located outside of the brain.  Where EEG is able to measure the electrical potential of the brain, typically through electrodes placed on the scalp, MEG measures very small changes in magnetic fields that are generated by the underlying electrical activity of clusters of neurons [George, Aine, Mosher, Schmidt, Ranken, Schlitt, Wood, Lewine, Sanders & Belliveau, 1995; Huettel, Song & McCarthy, 2009; Zillmer, Spiers & Culbertson, 2008].  As previously stated, communication between neurons is based on electrical fluctuations in membrane potential, and these fluctuations can be either excitatory (EPSP) or inhibitory (IPSP).  Because the skull and scalp are electrically and magnetically conductive, the field potentials generated by the brain can be recorded from arrays of electrodes placed on the scalp.  However, if we were to simply put an electrode on a persons scalp we would encounter a series of specific problems when attempting to measure localized changes in electrical potential between the neurons in the brain [for complete review see Huettel et al., 2009].  Although EEG and MEG signals originate from the same neurophysiological processes, there are a number of critical distinctions that can be made between the two processes [for complete review see Cohen & Cuffin, 1983]: first, magnetic fields are subject to less interferences from the skull and scalp versus electrical fields; second, EEG is sensitive to both radial and tangential components of current source whereas MEG is only sensitive to tangential components [suggesting that MEG only measures sulci selectively whereas EEG measures activity both in the sulci and at the top of the cortical gyri; see Brown, 2011]; third, from the second point it can be inferred that EEG is sensitive to activity in a wider range of brain areas, but the activity that is visible in MEG can be localized with higher accuracy [see Brown, 2011].  And finally, there are critical differences in the signal reception in EEG versus that of MEG: a magnetic field is typically generated as a response to a moving charge, while an electric field is generally due to a charge with no net movement [see Bloomfield, 2010].

As elaborated by Huettel et al. (2009), measuring changes in electrical potential can be inherently problematic for a number of reasons: (1) individual postsynaptic potentials (PSPs[1]) can last for a much longer duration than a typical action potential; (2) thousands of synapses make contacts with the dendrites, axons, and soma of a single neuron; (3) the volume current associated with an individual PSP extends into the extracellular space with a strength that diminishes rapidly as a function of distance from the source; and (4) the volume currents associated with individual PSPs combine in the extracellular space.  Ultimately, the signal recorded by an electrode will reflect the sum (å) of all PSPs produced at every synapse of every neuron in the brain, weighted according to their individual distance from the electrode.  Thus if an electrode is placed on the scalp, it is going to receive electrical data from groups of neurons so large one would not be able to differentiate target activity (i.e., signal) over that of the summated oscillations (i.e., noise) generated by the whole network.  This difficulty in signal localization, in both electrical and magnetic fields, is known as the inverse problem[2].  However, due to the quadratic decay of electrical/magnetic field strength as a function of distance, a given dipole will have very different effects on proximal electrodes when compared to electrodes that are located more distally [for review see Huettel et al., 2009; Zamrini, Maestu, Pekkonen, Funke, Makela, Riley, Bajo, Sudre, Fernandez, Castellanos, Pozo, Stam, Van Dijk, Bagic & Becker, 2011].  The ability to localize target signals over the systemic noise comes from the averaging of many trials.  This averaging eliminates local field fluctuations that are not related to the specific event of interest leaving only those signal-averaged field potentials that are time-locked to a stimulus [i.e., evoked potentials[3]; see Huettel et al., 2009].

1.2   Sidestepping the Inverse Problem: SQUIDs

As reviewed by Zamrini et al. (2011), the magnetic signals of the brain vary between 102 (evoked cortical activity) and 103 (the human a-rhythm) femtoteslas (fT; 10-15 T), whereas the magnetic field of the earth is approximately 109 fT [see also Huettel et al., 2009].  This leads to two problems: (1) the weakness of the signal itself, and (2) the amount of signal being generated over-and-above that of general neural activity.  The obvious question that arises is how do we detect these extremely small magnetic fluctuations?  When electricity flows through certain metals, such as copper (29Cu) wire, at room temperature there is always some resistance to the electrons flowing through the metal.  Some metals are able to conduct electricity without any resistance, and are called superconducting-metals [see Range, 2004].  The property of superconductivity is achieved by MEG uses highly sensitive magnetometers called Superconducting QUantum Interference Devices (SQUIDs) to measure the extremely weak magnetic fields.  These SQUIDs are typically comprised of pure niobium (41Nb), or a lead (82Pb) alloy with up to 10% gold (79Au) or indium (49In), as pure lead (82Pb) is highly unstable at the low temperatures that are required to maintain superconductivity [see Drung, Abmann, Beyer, Kirste, Peters, Ruede & Schurig, 2007].  In fact liquid helium (helium-4; ℓHe) exists at approximately 5.2 °K (-267.95 °C; -450.31 °F) [Eshraghi, Sasada, Kim & Lee, 2009].  Much like the distribution of electrodes in EEG recordings, modern MEG systems use large arrays of SQUID sensors in order to obtain scalp-recorded magnetic field oscillations.  As previously mentioned, SQUIDs are highly sensitive.  For example, SQUIDs have the ability to measure magnetic fields as low as 5 atto-tesla (aT; 5 x 10-18 T) [Range, 2004].

  1. 2.     Current Applications of Magnetoencephalography

MEG is currently developing use in a number of diverse clinical populations, as one of its primary advantages over other techniques is its superior temporal resolution.  MEG is able to reliably record from populations of neurons on the millisecond time scale [for examples see Kuriki, Sadamoto & Takeda, 2005; Nishitani, Avikainen & Hari, 2004; Shiraishi, 2011; Zamrini et al., 2011].  Additionally, the estimated localization of electromagnetic activity can be superimposed on an anatomic image of the brain, typically in the form of a magnetic resonance imaging (MRI) scan, to produce a more functionally based anatomical image of the brain [see Nishitani et al., 2004].

2.1   Research, Diagnosis, and Treatment of Epilepsy Using MEG

Epilepsy is a pervasive neurological disorder that is characterized by disruptions in the cortical and subcortical equilibriums of two common neurotransmitters: glutimate and g-aminobutyric acid (GABA).  The over-excitation caused by epileptic seizure activity can result in progressive neurological deterioration if left untreated [see Helmstaedter, Kurthen, Lux, Reuber & Elger, 2003; Piazzini, Turner, Chifari, Morabito, Canger & Canevini, 2006].  A growing body of literature describes the application of MEG in the clinical investigation of epileptic patients [Cappell, Schevon & Emerson, 2006; Knake, Grant, Stufflebeam, Wald, Shiraishi, Rosenow, Schomer, Fischl, Dale & Halgren, 2004; Mäkelä, Forss, Jääskeläinen, Kirveskari, Korvenoja & Paetau, 2006; Schwartz, Dlugos, Storm, Dell, Magee, Flynn, Zarnow, Zimmerman & Roberts, 2008; Shibasaki, Ikeda & Nagamine, 2007; Shiraishi, 2011; Rampp & Stefan, 2007].  Specifically, MEG could potentially be a highly advantageous technique when examining the source localization of epileptogenic foci as well as defining excitotoxic neocortical lesions in candidates of neurosurgical intervention [see Nakasato, Levesque, Barth, Baumgartner, Rogers & Sutherling, 1994; Shiraishi, 2011; Shiraishi, Watanabe, Watanabe, Inoue, Fujiwara & Yagi, 2001; Sutherling, Crandall, Engel, Darcey, Cahan & Barth, 1987].  The obvious goal of neurosurgical intervention in clinical epilepsy is to remove to epileptogenic tissue, while sparing as much healthy brain tissue as possible [see Luders, 1992].

Current literature suggests that MEG could potentially localize epileptogenic foci through the detection of abnormal interictal[4] electromagnetic brain activity.  Additionally, this technique has been shown to be extremely useful when anatomical lesions cannot be located using tradition magnetic resonance imaging [RamachandranNair, Otsubo, Shroff, Ochi, Weiss, Rutka & Snead, 2007].  As discussed by Shiraishi (2011), the primary advantage, as well as the current use, of MEG in epileptic populations comes from the diagnostic specification of epileptic foci prior to neocortical epileptic surgery.  It has also been shown that the high temporal resolution of MEG, coupled with the high temporal resolution of something like magnetic resonance imaging (MRI) can be used in concert to localize seizure activity.  Shiraishi (2011) reported using an MEG-guided MRI scan to generate functional/anatomical images in children to localize seizure activity.  Once the MEG-MRI has been coordinated, the source locations can be transformed to a 3D representation using something like a Talairach’s standard [see Nishitani & Hari, 2000; Nishitani et al., 2004; Talairach & Tournoux, 1988].  As will be discussed further, this is incredibly advantageous in that the MEG-MRI yields very good temporal and spatial resolution.

2.2   The Use of MEG in the Research of Visual Processing

Numerous magnetoencephalographic (MEG) studies have investigated the functional organization of the human visual processing system.  Cohen (1968) made the first electromagnetic recordings associated with a-rhythms using a single SQUID.  Over-and-above those imaging methods that are able to provide more superior spatial resolution [e.g., functional magnetic resonance imaging (fMRI); positron emission tomography (PET)], the temporal resolution provided by MEG makes it an excellent option to study functional changes in the human visual system.  In fact numerous studies using MEG [see chapter in Aine & Stephen, 2003], and EEG [see also Proverbio, Del Zotto & Zani, 2010] have examined the underlying structural components of the visual processing system.

2.2.1. MEG and Visual Processing

  As reviewed by Aine and Stephen (2003), in the human visual processing system, visual information flows from the retinas to the cortex through two primary pathways: (1) the tectopulvinar system (in primates) projects from the retina to the superior colliculus, or lateral posterior nucleus of the thalamus, to extrastriate areas such as parietal cortex [see also Wilson, 1978]; and (2) the geniculostriate system, which projects from the retina to the lateral geniculate nucleus (LGN) of the thalamus, and from the LGN to layer IV of visual cortical area 1 (V1; the primary visual cortex located in the occipital lobe) [see also Felleman & Van Essen, 1991].  V1 then projects to large portions of temporal and parietal cortices [for review see Aine & Stephen, 2003].  Many of the visual areas contained within these respective regions are organized in a topographical manner of the contralateral visual hemisphere and early studies using MEG experienced difficulty in the localization of visual processing because they failed to identify the structures/cell populations generating the magnetic fields measured at the scalp (i.e., early MEG studies did not superimpose source locations onto a technique with a higher spatial resolution, such as MRI) [see also Rowley & Roberts, 1995].

Hashimoto and colleagues (1999) used visual stimuli in the form of a checkerboard arrangement to attempt to find peaks in MEG data that would be similar to those found previously in studies using EEG data [Kurita-Tashima, Tobimatsu, Nakayama-Hiromatsu & Kato, 1991].  The results from Hashimoto et al. (1999) supported previous data using visually evoked potentials in that 3 distinct peaks were found (N75, P100, N145); additionally, the researchers attempted to localize the cortical epicenters of these individual peaks.  It was noted by the authors that the first (N75) and third (N145) peaks were adjacent to one another with orientations that were similar around the calcarine fissure.  However, the epicenter of the second peak (P100) was localized to the medial occipital lobe.  Of particular interest, and noted by Hashimoto et al. (1999), when different contrast levels were applied to the checkerboard arrangement the first peak (N45) was quite sensitive to contrast fluctuations whereas the second (P100) and third (N145) peaks remained relatively robust.  It was concluded that the first and third components had different physiological properties, indicating a recurrence of activation in both the striate and extrastriate cortices (located proximal to the calcarine fissure).  These data may suggest that the third component recorded by the MEG may have been a processing redundancy in primary visual cortex [V1; see also Kurita-Tashima et al., 1991].  The critical contribution of the study by Hashimoto and colleagues (1999) was the novel finding that the first (N45) and third (N145) components of visually evoked magnetic fields appear to originate from anatomically proximal, and almost identical sources, but their inherent physiological properties are distinct.

2.2.2 MEG and Higher Order Visual Processes

A critical process that directly parallels visual processing is the higher order cognitive processing of visual information.  In many cases, one could posit the argument that higher order processing is, in fact, predicated by visual/sensory processing.  Early studies using event-related potentials to measure the selective visual attention of humans seem to attempt to address the questions of when and where stimulus information processing is selected for further processing [reviewed in Aine & Stephen, 2003].  A recent study by Vanni and Uutela (2000) aimed to investigate the role of the frontoparietal network in visual attention using MEG.  It had been found previously that the frontoparietal network is active when there are reallocations of attention between objects or locations.  Vanni and Uutela (2000) compared precentral and parietal responses to peripheral stimuli when subjects were instructed to fixate on a small, centrally located, square.  At the same time subjects were asked to either attend or not attend to peripheral stimuli¾in the form of intermittent auditory tones.  The results demonstrated the most systematic differences between conditions was noted in the right precentral region.  The author’s interpretation of the data indicated that the focusing of attention on a fixation point enhances precentral cortical responses to stimuli in all parts of the visual field, whether attended to or not.  Additionally visually evoked responses, given at about 100 milliseconds, depend on whether visual attention is divided [Vanni & Uutela, 2000].  Put more simply, in conditions in which subjects were required to fixate centrally versus a control condition in which subjects were detecting tones, the precentral cortical region will be more active in subjects who are not involved in the distracting auditory paradigm.

2.2.3 MEG and Visual Processing in Asperger’s Syndrome

 Asperger’s syndrome is part of the Autism-spectrum disorders.  This class of disorders is primarily characterized by individuals experiencing moderate to severe impairments in social situations and situations that require interpersonal communication.  These deficits are often attributed to the possibility that Autistic individuals lack the ability to fully comprehend/understand the mental states of other people [see Caronna, Milunsky & Tager-Flusberg, 2008; Nishitani et al., 2004].  Recent research suggests that the functional integrity of the brains mirror neuron systems may play a role in the inability to function in social situation in individuals diagnosed with Autism or an Autism spectrum disorder (ASD).  These mirror neurons may also be exceedingly critical in a person’s ability to understand the intentions of other people [see Passingham, 1993; Rizzolatti, Fogasi & Gallese, 2001].  In a recent study by Nishitani and colleagues (2004) MEG was used in conjunction with MRI to examine any potential differences that may exist it the ability to detect and imitate facial gestures.  In the present study still pictures of three orofacial gestures were presented: (1) lip protrusion, (2) lip opening, and (3) bilateral contraction of the sides of the mouth.  As previously mentioned, the MEG data was couple with MRI in order to provide both adequate spatial and temporal resolution.  In the study by Nishitani et al. (2004), subjects’ averaged signals were digitally low-pass filtered at 40Hz. Additionally, the current sources of the spatial signal patterns were modeled as equivalent current dipoles (ECDs)[5].  These ECDs that best explained the most dominant signals were obtained and only those that accounted for greater than 80% of the field variance were further analyzed[6].  The ECDs were then superimposed on each subjects MRI surface rendering following alignment of the MEG-MRI coordinate systems [as reported in Hämäläinen, Hari, Ilmoniemi, Knuutila & Lounasmaa, 1993].  The results of this study demonstrated two important neurological processes: first, visual information processing follows an extremely linear route in the human visual system; specifically, the MEG signals of the control group and Asperger’s group from five distinct cortical areas in the left hemisphere: (1) occipical lobe for visual processing, (2) superior temporal sulcus for the recognition and processing of biologically active stimuli, (3) inferior parietal lobe for the perception of emotion in facial stimuli, (4) the mirror neurons located in inferior frontal cortex were then activated to prepare to emulate the action of another person, and (5) primary motor cortex generated the signals to apply the emulation.  The second critical point of this study was that activity in the inferior frontal cortex of individuals with Asperger’s was severely impaired relative to controls in ether receiving, processing, or transmitting information to motor cortex [see Nishitani et al., 2004].

2.2.4 Conclusions of Visual System Research Using MEG

  MEG has been shown to be highly useful in the research of the human visual processing system.  A number of studies have been shown to be highly successful when utilizing the temporal benefits of MEG [Hashimoto et al., 1999; Nishitani et al., 2004].  Additionally, when coupled with MRI, MEG can provide an excellent mixture of spatial and temporal resolution and may potentially serve as a useful diagnostic tool for neurological disorders that may fall outside of the scope of other, more standard, imaging techniques.

2.3 Using MEG to Localize Brain Tumors

The application of MEG to the localization of astrocytomas or gliomas is approximately similar in principal to the localization of epicenters of epileptic activity.  When removing a brain tumor the primary goal of a neurosurgeon, just like the goal of a surgical intervention in a patient with epilepsy, is to remove as much of the diseased part of the brain as possible while attempting to spare the healthy brain tissue.  The relationship between abnormal areas that should be targeted for surgical excision, the sparing of proximally located functional brain tissue and the potential for operational success are critical considerations prior to tumor excision.  The current and continued use of MEG in the diagnostics of neurological disorders and diseases as been shown to be highly practical, the mapping of brain activity can be very efficient at spotlighting areas that are not functioning adequately.  When coupled with a high resolution MRI, MEG has been shown to be an excellent, non-invasive diagnostic tool that brings anatomical and physiological data together to yield functional images of the brain with high temporal and spatial resolution.

2.3.1 Advantages of MEG-MRI Assisted Craniotomy

Computer assisted neurosurgery provides a large number of advantages to the surgeon.  First and foremost, using MEG coupled with MRI in order to gain a 3-dimensional perspective of the tumor allows the surgeon to better understand the size and range, but more importantly this perspective helps the surgeon to obtain a better understanding of the specific vasculature in the area surrounding the focal point of the tumor[7].  The increased spatial and temporal resolution also allows for a more specific excision of unhealthy brain tissue with less risk to surrounding cortical and vascular anatomy.

  1. 3.     Conclusions and Future Directions

Although MEG has a number of distinct limitations, the current literature seems to suggest that the supportive role that this technique plays in combination with other techniques such as EEG and MRI will continue to play a critical role in the way that we diagnose, and potentially treat, a number of serious neurological deficits and disease states.




Aine, C.J., & Stephen, J.M. (2003). MEG studies of visual processing. In The cognitive electrophysiology of mind and brain (Ch. V, pp. 93-142), Edited by: Zani A, Proverbio AM. Academic Press/Elsevier

Azevedo, F.A., Carvalho, L.R., Grinberg, L.T., Farfel, J.M., Ferretti, R.E., Leite, R.E., Filho, W.J., Lent R., & Herculano-Houzel, S. (2009). Equal numbers of neuronal and non-neuronal cells make the human an isometrically scaled-up primate brain. The Journal of Comparative Neurology, 513:532-541.

Bloomfield, L.A. (2010). How things work: The physics of everyday life (4th Ed). © 2010, John Wiley & Sons, Inc.

Brown, T. (April, 2011). Multimodal MEG: Methods and developmental applications. SDSU-UCSD lecture series.

Cappell, J., Schevon, C., & Emerson, R.G. (2006). Magnetoencephalography in epilepsy: tailoring interpretation and making inferences. Current Neurology and Neuroscience Reports, 6:327-331.

Caronna, E.B., Milunsky, J.M., & Tager-Flusberg, H. (2008). Autism spectrum disorders: clinical and research frontiers. Archives of Disease in Childhood, 93(6):518-523.

Chu-Shore, C.J., Kramer, M.A., Bianchi, M.T., Caviness, V.S., & Cash, S.S. (2011). Network Analysis: Applications for the Developing Brain. Journal of Child Neurology, 26(4):488-500.

Cohen, D. (1968). Magnetoencephalography: evidence of magnetic fields produced by alpha-rhythm currents. Science, 161:784-786.

Cohen, D., & Cuffin, B.N. (1983). Demonstration of useful differences between the magnetoencephalogram and elecroencephalogram. Electroencephalography and Clinical Neurophysiology, 56:38-51.

Drachman, D. (2005). Dow we have brain to spare? Neurology, 64(12):2004-2005.

Drung, D., Abmann, C., Beyer, J., Kirste, A., Peters, M., Ruede, F., & Schurig, Th. (2007). Highly sensitive and easy-to-use SQUID sensors. Applied superconductivity conference, Seattle, WA. vol. 17 (1), n˚2 (p. 927).

Eshraghi, M.J., Sasada, I., Kim, J.M., & Lee, Y.H. (2009). AIN ceramic thermal guide for cooling a SQUID and its effect on thermal magnetic noise. Engineering Sciences Reports, Kyushu University, Vol.31,No.2,pp.1-4.

Felleman, D.J., & Van Essen, D.C. (1991). Distributed hierarchical processing in the primate cerebral cortex. Cerebral Cortex, 1:1-47.

George, J.S., Aine, C.J., Mosher, J.C., Schmidt, D.M., Ranken, D.M., Schlitt, H.A., Wood, C.C., Lewine, J.D., Sanders, J.A., & Belliveau, J.W. (1995). Mapping function in the human brain with magnetoencephalography, anatomical magnetic resonance imaging, and functional magnetic resonance imaging.  Journal of Clinical Neurophysiology, 12(5):406-431.

Hämäläinen, M., Hari, R., Ilmoniemi, R.J., Knuutila, J., & Lounasmaa, O.V. (1993). Magnetoencephalography—theory, instrumentation, and applications to noninvasive studies of the working human brain. Reviews of Modern Physics, 65:413-497.

Hashimoto, I., Kashii, S., Kikuchi, M., Honda, Y., Nagamine, T., & Shibasaki, H. (1999). Temporal profile of visual evoked responses to pattern-reversal stimulation analyzed with a whole-head magnetometer. Experimental Brain Research, 125:375-382.

Helmstaedter, C., Kurthen, M., Lux, S., Reuber, M., & Elger, C.E. (2003). Chronic epilepsy and cognition: A longitudinal study in temporal lobe epilepsy. Annals of Neurology, 54(4):425-432.

Huettel, S.A., Song, A.W., & McCarthy, G. (2009). Functional Magnetic Resonance Imaging, 2nd Edition. (Ch. I, pp. 8; Ch. VI, pp. 159; Ch XIII, pp. 464-474) Copyright © 2009 by Sinauer Associates, Inc.

Huttenlocher, P.R., & Dabholkar, A.S. (1997). Regional differences in synaptogenesis in human cerebral cortex. Journal of Comparative Neurology, 387:167-178.

Knake, S., Grant, E.P., Stufflebeam, S.M., Wald, L.L., Shiraishi, H., Rosenow, F., Schomer, D.L., Fischl, B., Dale, A.M., & Halgren, E. (2004). Aids to telemetry in the presurgical evaluation of epilepsy patients: MRI, MEG and other non-invasive imaging techniques. Advances in Clinical Neurophysiology (Supplements to Clinical Neurophysiology, Vol. 57). 57:494-502.

Kuriki, I., Sadamoto, K., & Takeda, T. (2005). MEG recording from the human ventro-occipital cortex in response to isoluminant color stimulation. Visual Neuroscience, 22(3):283-293.

Kurita-Tashima, S., Tobimatsu, S., Nakayama-Hiromatsu, M., & Kato, M. (1991). Effect of check size on the pattern reversal visual evoked potential. Electroencephalography and Clinical Neurophysiology/Evoked Potentials Section, 80(3):161-166.

Luders, H.O. (1992). Epilepsy Surgery, New York Raven Press © 1992.

Mäkelä, J.P., Forss, N., Jääskeläinen, J., Kirveskari, E., Korvenoja, A., & Paetau, R. (2006). Magnetoencephalography in neurosurgery. Neurosurgery, 59:493-510.

Nakasato, N., Levesque, M.F., Barth, D.S., Baumgartner, C., Rogers, R.L., & Sutherling, W.W. (1994). Comparisons of MEG, EEG, and ECoG source localization in neocortical partial epilepsy in humans. Electroencephalogrophy in Clinical Neurophysiology, 91:171-178.

Nishitani, N., Avikainen, S., & Hari, R. (2004). Abnormal imitation-related cortical activation sequences in Asperger’s syndrome. Annals of Neurology, 55:558-562.

Nishitani, N., & Hari, R. (2000). Temporal dynamics of cortical representation for action. Proceedings of the National Academy of Sciences: United States of America, 97:913-918.

Nishiyama, A., Yang, Z., & Butt, A. (2005). Astrocytes and NG2-glia: what’s in a name? Journal of Anatomy, 207:687-693.

Pakkenberg, B., Pelvig, D., Marner, L., Bundgaard, M.J., Gunderson, H.J., Nyengaard, J.R., & Regeur, L. (2003). Aging and the human neocortex. Experimental Gerentology, 38:95-99.

Passingham, R. (1993). The frontal lobe and voluntary action. Oxford: Oxford University Press, © 1993.

Piazzini, A., Turner, K., Chifari, R., Morabito, A., Canger, R., & Canevini, M.P. (2006). Attention and psychomotor speed decline in patients with temporal lobe epilepsy: a longitudinal study. Epilepsy Research, 72(2-3):89-96.

Proverbio, A.M., Del Zotto, M., & Zani, A. (2010). Electrical neuroimaging evidence that spatial frequence-based selective attention affects V1 activity as early as 40-60 ms in humans. BMC Neuroscience, 11:59-72.

RamachandranNair, R., Otsubo, H., Shroff, M.M., Ochi, A., Weiss, S.K., Rutka, J.T., & Snead, O.C. (2007). MEG predicts outcome following surgery for intractable epilepsy in children with normal or nonfocal MRI findings. Epilepsia, 48(1):149-157.

Rampp, S., & Stefan, H. (2007). Magnetoencephalography in presurgical epilepsy diagnosis. Expert Reviews – Expert Review of Medical Devices, 4:335-347.

Range, S. (2004). GRAVITY PROBE B: Examining Einstein’s Spacetime with Gyroscopes. © National Aeronautics and Space Administration. Retrieved From:

Rizzolatti, G., Fogasi, L., & Gallese, V. (2001). Neurophysiological mechanisms underlying the understanding and imitation of action. Nature Reviews Neuroscience, 2:661-670.

Rowley, H.A., & Roberts, T.P. (1995). Functional localization by magnetoencephalography. Neuroimaging Clinics of North America, 5:695-710.

Schwartz, E.S., Dlugos, D.J., Storm, P.B., Dell, J., Magee, R., Flynn, T.P., Zarnow, D.M., Zimmerman, R.A., & Roberts, T.P.L. (2008). Magnetoencephalography for pediatric epilepsy: How we do it. American Journal of Neuroradiology, 29:832-837.

Shiraishi, H. (2011). Source localization in magnetoencephalography to identify epileptogenic foci. Brain & Development, 33:276-281.

Shiraishi, H., Watanabe, Y., Watanabe, M., Inoue, Y., Fujuwara, T., & Yagi, K. (2001). Interictal and ictal magnetoencephalography study in patients with medial frontal lobe epilepsy. Epilepsia, 47:355-364.

Sutherling, W.W., Crandall, P.H., Engel, J., Darcey, T.M., Cahan, L.D., & Barth, D.S. (1987). The magnetic field of complex partial seizure agrees with intracranial localizations. Annuals of Neurology, 21:548-558.

Talairach, J., & Tournoux, P. (1988). Co-planar stereotaxic atlas of the human brain 3-dimensional proportional system: an approach to cerebral imaging.  Stuttgart: Thieme, 1988.

Vanni, S., & Uutela, K. (2000). Foveal attention modulates responses to peripheral stimuli. Journal of Neurophysiology, 83:2443-2452.

Wilson, M. (1978). Visual System: Pulvinar-extrastriate cortex. In R.B. Masterton (Ed) Handbook of Behavioral Neurobiology, (Vol. 1, pp. 209-247), Plenum, New York.

Zamrini, E., Maestu, F., Pekkonen, E., Funke, M., Makela, J., Riley, M., Bajo, R., Sudre, G., Fernandez, A., Castellanos, N., del Pozo, F., Stam, C.J., Van Dijk, B.W., Bagic, A., & Becker, J.T. (2011). Magnetoencephalography as a putative biomarker for Alzheimer’s disease. International Journal of Alzheimer’s Disease, Article ID 280289, doi:10.4061/2011/280289

Zillmer, E.A., Spiers, M.V., & Culbertson, W.C. (2008). Principles of Neuropsychology, Second Edition. (Ch. II, pp. 56), © 2008, 2001 Thomson Wadsworth.

[1] Any postsynaptic potential, excitatory or inhibitory, that results from synaptic activity.

[2] The mathematical impossibility of determining the distribution of electrical sources within an object based on the measurement of electrical or magnetic fields at the surface of the object [Huettel et al., 2009].

[3] A field potential that occurs in response to a specific sensory stimulus [Huettel et al., 2009].

[4] Refers to the periods of time between seizure activity.

[5] A common source-modeling technique for MEG involves calculating a set of equivalent current dipoles, which assume the underlying neuronal sources to be focal.  This model fitting procedure is consistently non-linear and over-determined [which makes sense since the number of unknown dipole parameters is smaller than the number of MEG measurements].

([N(N-1)/2] = Non-Redundant Elements) {N = the number of dipole parameters}

*Based on the number of non-redundant covariances, we can estimate as many dipole parameters as we have MEG measurements in order to have a just-identified model (df = 0).

[6] Nishitani et al. (2004) only retain those ECDs that reflect a coherent neuronal focal source.

[7] Although MEG is not going to be directly measure hemodynamic response like a functional MRI would, the large amount of temporal resolution may actually allow for a better understanding of the nature of the vascularity of the tumor as changes in BOLD signal are subject to a greater time delay.


About dwmaasberg

Memories are physical connections between neurons. I think that is pretty cool!
This entry was posted in Electrophysiology, Imaging, Methods, Neuroscience and tagged . Bookmark the permalink.

1 Response to Magnetoencephalography: Clinical Uses and Limitations

  1. jzulli says:

    Fascinating stuff as always. Keep up your posts! Even non-scientists (like myself) are interested in this kind of blogging.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s