Magnetoencephalography with optically pumped magnetometers (OPM-MEG): the next generation of functional neuroimaging
2022; Elsevier BV; Volume: 45; Issue: 8 Linguagem: Inglês
10.1016/j.tins.2022.05.008
ISSN1878-108X
AutoresMatthew J. Brookes, James Leggett, Molly Rea, Ryan M. Hill, Niall Holmes, Elena Boto, Richard Bowtell,
Tópico(s)Advanced MRI Techniques and Applications
ResumoMagnetoencephalography (MEG) allows noninvasive electrophysiological imaging of human brain activity. However, current MEG technology has significant limitations.Optically pumped magnetometers (OPM)-MEG is a new type of MEG instrumentation, promising several advantages compared with conventional scanners: higher signal sensitivity, better spatial resolution, more uniform coverage, lifespan compliance, free movement of participants during scanning, and lower system complexity.We describe the principles underlying OPM-MEG and its components, including noncryogenic field sensors and magnetic shielding technologies.We discuss how the OPM-MEG technology is impacting neuroscience, enabling researchers to overcome limitations of conventional human imaging techniques and tackle new types of research questions. Magnetoencephalography (MEG) measures human brain function via assessment of the magnetic fields generated by electrical activity in neurons. Despite providing high-quality spatiotemporal maps of electrophysiological activity, current MEG instrumentation is limited by cumbersome field sensing technologies, resulting in major barriers to utility. Here, we review a new generation of MEG technology that is beginning to lift many of these barriers. By exploiting quantum sensors, known as optically pumped magnetometers (OPMs), 'OPM-MEG' has the potential to dramatically outperform the current state of the art, promising enhanced data quality (better sensitivity and spatial resolution), adaptability to any head size/shape (from babies to adults), motion robustness (participants can move freely during scanning), and a less complex imaging platform (without reliance on cryogenics). We discuss the current state of this emerging technique and describe its far-reaching implications for neuroscience. Magnetoencephalography (MEG) measures human brain function via assessment of the magnetic fields generated by electrical activity in neurons. Despite providing high-quality spatiotemporal maps of electrophysiological activity, current MEG instrumentation is limited by cumbersome field sensing technologies, resulting in major barriers to utility. Here, we review a new generation of MEG technology that is beginning to lift many of these barriers. By exploiting quantum sensors, known as optically pumped magnetometers (OPMs), 'OPM-MEG' has the potential to dramatically outperform the current state of the art, promising enhanced data quality (better sensitivity and spatial resolution), adaptability to any head size/shape (from babies to adults), motion robustness (participants can move freely during scanning), and a less complex imaging platform (without reliance on cryogenics). We discuss the current state of this emerging technique and describe its far-reaching implications for neuroscience. Brain imaging via MEGMEG [1.Cohen D. Magnetoencephalography: evidence of magnetic fields produced by alpha-rhythm currents.Science. 1968; 161: 784-786Crossref PubMed Scopus (439) Google Scholar] is a noninvasive method allowing real-time imaging of brain function. The technique is based on measurement of magnetic fields outside the head, which are generated (primarily) by synchronous dendritic current flow through neuronal assemblies. Mathematical modelling of these fields enables the generation of 3D images (termed source localisation) showing how electrical activity changes, moment-to-moment, as the brain responds to various experimental scenarios or cognitive demands. MEG has temporal resolution in the millisecond range and spatial resolution of ~2–5 mm [2.Bonaiuto J.J. et al.Non-invasive laminar inference with MEG: comparison of methods and source inversion algorithms.NeuroImage. 2018; 167: 372-383Crossref PubMed Scopus (22) Google Scholar]. With these characteristics, MEG has many advantages over other functional imaging modalities [3.Baillet S. Magnetoencephalography for brain electrophysiology and imaging.Nat. Neurosci. 2017; 20: 327-339Crossref PubMed Scopus (350) Google Scholar], including functional magnetic resonance imaging (fMRI), which is limited to haemodynamic metrics and has limited temporal resolution, and electroencephalography (EEG), the spatial resolution of which is limited by distortions in electrical potential caused by the skull. Consequently, MEG has become an important part of the neuroscientific toolbox for noninvasive imaging. However, the present generation of MEG scanners has significant limitations, which hamper their utility.The fundamental problem limiting MEG's applicability is that, to gain sufficient sensitivity to measure the small (~100 fT) magnetic fields generated by the brain, current MEG systems (which are housed in a magnetically shielded environment to suppress background fields) employ pick-up coils that are coupled to superconducting quantum interference devices (SQUIDs) [4.Cohen D. Magnetoencephalography: detection of the brain's electrical activity with a superconducting magnetometer.Science. 1972; 5: 664-666Crossref Scopus (469) Google Scholar,5.Hamalainen M.S. et al.Magnetoencephalography: theory, instrumentation, and applications to non-invasive studies of the working human brain.Rev. Mod. Phys. 1993; 65: 413-497Crossref Scopus (3502) Google Scholar]. These sensors typically require cooling to ~4 K (–269°C). This, in turn, means sensors are bathed in liquid helium and a vacuum is maintained between the sensors and the participant's scalp, for thermal insulation. Sensors must consequently be formed into a fixed array around the scalp. These design considerations underlie many of the limitations of MEG. First, the fixed array means that participants must remain still relative to the sensors throughout data acquisition. Coping with the MEG environment is consequently challenging for some participants. Second, the MEG signal strength decreases with the square of distance from the source (the inverse square law); the requirement for thermal insulation between the scalp and the sensor limits proximity (the closest a sensor can be to the scalp is ~2 cm) and this limits signal strength. The need for rigidity also means that a 'one size fits all' MEG helmet, built to fit ~95% of adults, is used. In practice, this means that the helmet is designed for someone with a relatively large head; most people will not fit the helmet perfectly and the gap between the scalp and the helmet will vary across the head. This results in inhomogeneous coverage. For those with smaller heads this effect is amplified and it is hard to simultaneously obtain uniform coverage and high sensitivity in infants. Finally, the complexities of the system make scanners costly to buy and maintain and the need for cryogenics means either a constant supply of liquid helium or a local helium reliquefier is needed.In recent years, the MEG field has seen the introduction of a new magnetic field-sensing technology. OPMs [6.Allred J. et al.High-sensitivity atomic magnetometer unaffected by spin-exchange relaxation.Phys. Rev. Lett. 2002; 89130801Crossref PubMed Google Scholar] are magnetic-field sensors that offer sensitivity comparable to SQUIDs without relying on cryogenic cooling. This has resulted in the evolution of new MEG systems (e.g., [7.Johnson C.N. et al.Multi-sensor magnetoencephalography with atomic magnetometers.Phys. Med. Biol. 2013; 58: 6065-6077Crossref PubMed Scopus (82) Google Scholar]) and, though still nascent technology, 'OPM-MEG' scanners are beginning to out-perform the current state of the art, offering higher quality data, improved uniformity of coverage, motion robustness, and lower system complexity. In this review, we outline the current state of OPM-MEG, describing the technology, its advantages, and limitations. We also review current literature and speculate on where this new technology could take the neuroimaging field.The technical advantages of OPMsOPMs use the quantum properties of atoms to sense local magnetic fields [8.Tierney T.M. et al.Optically pumped magnetometers: from quantum origins to multi-channel magnetoencephalography.NeuroImage. 2019; 199: 598-608Crossref PubMed Scopus (82) Google Scholar]. OPMs have been in development for several decades, with recent years seeing marked improvements in sensitivity and miniaturisation. The devices that have become popular for MEG are small, self-contained units, approximately the size and shape of a (2 × 4) Lego brick (Figure 1A ) [9.Shah V.K. Wakai R.T. A compact, high performance atomic magnetometer for biomedical applications.Phys. Med. Biol. 2013; 58: 8153-8161Crossref PubMed Scopus (150) Google Scholar, 10.Johnson C. et al.Magnetoencephalography with a two-color pump-probe, fiber-coupled atomic magnetometer.Appl. Phys. Lett. 2010; 97243703Crossref Scopus (132) Google Scholar, 11.Sander T.H. et al.Magnetoencephalography with a chip-scale atomic magnetometer.Biomed. Opt. Express. 2012; 3: 981-990Crossref PubMed Scopus (206) Google Scholar]. Each unit comprises a glass cell containing an atomic vapour (alkali atoms, usually 87Rb), a laser and associated optics to project polarised laser light through the cell, a set of electromagnetic coils for field control within the cell, and a photodiode for detection of light passing through the vapour (Figure 1B). With the atoms 'pumped' by the laser into a specific quantum state, the atomic vapour becomes magnetised and interacts with any external magnetic field (e.g., the neuromagnetic field) that passes through the sensor. Such interactions modulate the amount of light passing through the vapour and the field magnitude can be inferred by measurement at the photodiode (Box 1). Magnetic fields can be measured in two perpendicular orientations (in the plane perpendicular to the laser beam; Figure 1B) with a noise floor of around 7–10 fT/sqrt (Hz). [For comparison, most SQUIDs have a noise floor of around 2–5 fT/sqrt (Hz).]Box 1OPM physicsAt the heart of an OPM is a cell containing an atomic vapour, typically 87Rb (Figure I). In physics terms, the Rb atom possesses spin and hence has a magnetic moment. In the absence of external influence, the magnetic moments of the Rb atoms are aligned arbitrarily (shown by the black arrows in the upper panel of Figure I). However, if circularly polarised laser light, at a wavelength resonant with the D1 transition between quantum states (795 nm) is introduced, photon absorption causes a change in atomic energy state and an alignment of magnetic moments for all atoms in the vapour. The vapour essentially becomes magnetised, with a bulk magnetisation along the direction of the beam (Figure I, centre panel) [64.Happer W. Optical pumping.Rev. Mod. Phys. 1972; 44: 169Crossref Scopus (1488) Google Scholar]. Once in this state, the atoms can no longer absorb photons and the vapour becomes transparent to the laser light. This means the light measured at the photodiode is maximised (a 'zero-field resonance').However, the bulk magnetisation of the Rb atoms in the vapour obeys the Bloch equations and so, if a magnetic field impinges on the cell (e.g., the field from the brain) the bulk magnetisation undergoes Larmor precession (Figure I, bottom panel). This reduces the net alignment of the magnetic moments and the atoms can again absorb laser light, causing a drop in light intensity at the photodetector. Light intensity consequently becomes a (Lorentzian) function of (the transverse) magnetic field in the cell [65.Tannoudji C. et al.Diverses résonances de croisement de niveaux sur des atomes pompés optiquement en champ nul i. théorie.Rev. Phys. Appl. 1970; 5 (In French): 95Crossref Google Scholar].Figure ISchematic diagram of the operation of an optically pumped magnetometer (OPM).Show full captionLaser light is passed through a glass cell containing an Rb vapour onto a photo detector. Interaction of the light with the vapour causes the amount of light passing through the vapour to become a sensitive marker of magnetic field.View Large Image Figure ViewerDownload Hi-res image Download (PPT)OPMs of this type usually operate in the spin-exchange relaxation-free (SERF) regime [6.Allred J. et al.High-sensitivity atomic magnetometer unaffected by spin-exchange relaxation.Phys. Rev. Lett. 2002; 89130801Crossref PubMed Google Scholar], which requires the 87Rb-vapour to be heated to ~150°C. However, thermal insulation (e.g., aerogel) allows the cell to be placed just a few millimetres from the scalp, compared with ~2 cm or more in cryogenic MEG. The reduced sensor-scalp separation has two effects (Figure 2). First, the magnitude of measured magnetic field vectors is larger when sensors are placed on the scalp surface, compared with when sensors are placed at the distance required for cryogenic MEG. Simulations [12.Boto E. et al.The benefits of atomic magnetometers for MEG: a simulation study.PLoS One. 2016; 11e0157655Crossref PubMed Scopus (82) Google Scholar,13.Iivanainen J. et al.Measuring MEG closer to the brain: performance of on-scalp sensor arrays.Neuroimage. 2017; 147: 542-553Crossref PubMed Scopus (105) Google Scholar] show that this effect provides a four- to fivefold signal enhancement in many cortical areas. As a result of the nonlinearity in the inverse square law, this advantage declines with depth (e.g., to a factor of ~2 for deeper cortical regions [12.Boto E. et al.The benefits of atomic magnetometers for MEG: a simulation study.PLoS One. 2016; 11e0157655Crossref PubMed Scopus (82) Google Scholar] and likely lower in subcortical structures). Nevertheless, there is potential for enhanced signal strength across the whole brain. If the noise floors of OPMs and SQUIDs were equal, this would result in a similar increase in signal-to-noise ratio (SNR) for OPM-based systems. In practice, the noise floor of OPMs currently remains higher than that of a SQUID, nevertheless increased SNR for cortical sources has been realised experimentally; for example, in a healthy adult, the SNR of evoked responses in the sensory cortex generated by median nerve stimulation was improved by a factor of ~2 using OPMs compared with SQUID measurements [14.Boto E. et al.A new generation of magnetoencephalography: room temperature measurements using optically-pumped magnetometers.NeuroImage. 2017; 149: 404-414Crossref PubMed Scopus (198) Google Scholar]. In deeper sources, the advantage of proximity is (at present) negated by the higher noise of an OPM. That said, OPMs have been used to image subcortical structures, notably the hippocampus [15.Barry D.M. et al.Imaging the human hippocampus with optically-pumped magnetometers.NeuroImage. 2019; 203116192Crossref PubMed Scopus (23) Google Scholar]. Second, the enhanced proximity of the sensors means that measured field patterns are more 'focal' (Figure 2). This allows better separation of field patterns arising from spatially separate current sources in the brain. Simulations [13.Iivanainen J. et al.Measuring MEG closer to the brain: performance of on-scalp sensor arrays.Neuroimage. 2017; 147: 542-553Crossref PubMed Scopus (105) Google Scholar] show that the correlations between field patterns generated by separate sources are reduced approximately threefold by moving sensors closer to the scalp. This should translate to improved spatial resolution [16.Boto E. et al.Moving magnetoencephalography towards real-world applications with a wearable system.Nature. 2018; 555: 657Crossref PubMed Scopus (452) Google Scholar]. In support of this, recent simulation work [17.Nugent A.C. et al.On-scalp magnetocorticography with optically pumped magnetometers: simulated performance in resolving simultaneous sources.NeuroImage. 2022; 2100093PubMed Google Scholar] showed that a densely packed OPM array over a patch of cortex can localise electrophysiological brain responses at unprecedented resolution for a noninvasive device (a process the authors term 'magnetocorticography'). However, careful system calibration was a significant factor. Again, it is important to note that this is a function of depth, with the largest gains in superficial cortical areas (i.e., close to the skull). In sum, OPM-MEG offers two fundamental advantages in performance over current MEG: higher sensitivity and enhanced spatial resolution. Both effects are notable in all subjects, but are particularly pronounced when imaging individuals with smaller heads (e.g., infants).Figure 2Advantages of optically pumped magnetometer (OPM)-magnetoencephalography (MEG) compared with conventional MEG.Show full caption(A) A schematic representation of conventional MEG [superconducting quantum interference device (SQUID)-MEG]. A participant sits with their head in a static helmet (see inset photo, adapted from [16.Boto E. et al.Moving magnetoencephalography towards real-world applications with a wearable system.Nature. 2018; 555: 657Crossref PubMed Scopus (452) Google Scholar]), containing an array of field sensors (blue circles). Sensors require cryogenic cooling and are consequently bathed in liquid helium. The requirement for thermal insulation (provided by a vacuum, shown in grey) limits sensor proximity to the head, hence the size of the measured magnetic field (represented by the length of the black lines) is limited. For participants with smaller heads (and particularly infants or babies) the sensors would be even further away and, consequently, the signal-to-noise ratio even lower. (B) A schematic representation of OPM-MEG. OPMs do not require cryogenic cooling and so can be mounted flexibly in a lightweight helmet (see inset photo, adapted from [54.Hill R.M. et al.Multi-channel whole-head OPM-MEG: Helmet design and a comparison with a conventional system.NeuroImage. 2020; 219116995Crossref PubMed Scopus (67) Google Scholar]) that can be made to fit any head shape. Because sensors are closer to the head compared with conventional MEG, the measured fields (black lines) are larger, increasing sensitivity. In addition, closer proximity allows denser sampling of focal field patterns (examples of which are also shown in the insets), which increases spatial resolution.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Apart from permitting greater proximity to the scalp, OPMs also allow the design of bespoke sensor arrays; this means that the array can be tailored to best match the participant, or the experiment that is to be carried out. For example, helmets housing the sensors can come in different sizes [18.Hill R.M. et al.A tool for functional brain imaging with lifespan compliance.Nat. Commun. 2019; 10: 4785Crossref PubMed Scopus (51) Google Scholar], or can be made bespoke for a single participant [14.Boto E. et al.A new generation of magnetoencephalography: room temperature measurements using optically-pumped magnetometers.NeuroImage. 2017; 149: 404-414Crossref PubMed Scopus (198) Google Scholar], thus ensuring uniform coverage and adaptability to scan almost anyone, from babies to adults. Arrays can be designed to target specific brain regions with high sensor density, for example, if high spatial resolution is desired in a specific area (recently published examples include the language network [19.Tierney T.M. et al.Cognitive neuroscience using wearable magnetometer arrays: non-invasive assessment of language function.NeuroImage. 2018; 181: 513-520Crossref PubMed Scopus (31) Google Scholar], hippocampus [20.Tierney T.M. et al.Mouth magnetoencephalography: a unique perspective on the human hippocampus.NeuroImage. 2021; 225117443Crossref PubMed Scopus (17) Google Scholar], and cerebellum [21.Lin C.-H. et al.Using optically-pumped magnetometers to measure magnetoencephalographic signals in the human cerebellum.J. Physiol. 2019; 597: 4309-4324Crossref PubMed Scopus (14) Google Scholar]). It is also becoming apparent that OPM use is not limited to the brain, with arrays also having been used to measure electrophysiological signals in the muscles [22.Marquetand J. et al.Optically pumped magnetometers reveal fasciculations non-invasively.Clin. Neurophysiol. 2021; 132: 2681-2684Crossref PubMed Scopus (5) Google Scholar], peripheral nerves [23.Bu Y. et al.Peripheral nerve magnetoneurography with optically pumped magnetometers.Front. Physiol. 2022; 13798376Crossref Scopus (1) Google Scholar], spinal cord [24.Mardell L.C. et al.Concurrent spinal and brain imaging with optically pumped magnetometers.bioRxiv. 2022; (Published online May 13, 2022)https://doi.org/10.1101/2022.05.12.491623Google Scholar], retina [25.Westner B.U. et al.Contactless measurements of retinal activity using optically pumped magnetometers.NeuroImage. 2021; 243118528Crossref PubMed Scopus (2) Google Scholar], and the foetus [26.Strand S. et al.Low-cost fetal magnetocardiography: a comparison of superconducting quantum interference device and optically pumped magnetometers.J. Am. Heart Assoc. 2019; 20e013436Google Scholar]. Another advantage is that, whereas SQUIDs typically measure the magnetic field in one orientation (usually radial to the scalp), OPMs can simultaneously measure components of the magnetic field vector along multiple directions [27.Shah, V., et al. Quspin Inc. Zero field parametric resonance magnetometer with triaxial sensitivity, US10775450B1Google Scholar]. The tangential components of the neuromagnetic field are smaller than the radial components [13.Iivanainen J. et al.Measuring MEG closer to the brain: performance of on-scalp sensor arrays.Neuroimage. 2017; 147: 542-553Crossref PubMed Scopus (105) Google Scholar,28.Marhl U. et al.Simulation study of different OPM-MEG measurement components.Sensors. 2022; 22: 3184Crossref PubMed Scopus (2) Google Scholar] but still contain useful information. Studies have shown that the addition of tangential components offers advantages when trying to differentiate fields from within the brain from those originating outside the head (i.e., interference) [29.Brookes M.J. et al.Theoretical advantages of a triaxial optically pumped magnetometer magnetoencephalography system.NeuroImage. 2021; 236118025Crossref PubMed Scopus (19) Google Scholar] and when the number of sensors is limited [28.Marhl U. et al.Simulation study of different OPM-MEG measurement components.Sensors. 2022; 22: 3184Crossref PubMed Scopus (2) Google Scholar]. Thus, the flexibility of OPMs to make bespoke arrays, and to offer multidimensional magnetic field metrics, is bringing about a change in MEG capabilities.Towards wearabilityOPMs reached sensitivities comparable with SQUIDs in the early years of the 21st century, with the first MEG measurements (of the auditory evoked response) published in 2006 [30.Xia H. et al.Magnetoencephalography with an atomic magnetometer.Appl. Phys. Lett. 2006; 89211104Crossref Scopus (309) Google Scholar]. However, the experimental set-up was laboratory-based and the large footprint of the OPMs negated the possibility of simultaneous measurements from a large number of sites on the scalp. Nevertheless, the potential was recognised and there followed a period of miniaturisation, with lightweight OPMs emerging around the early 2010s. By 2010, a small SERF magnetometer had been used to measure evoked responses from median nerve and auditory stimulation [10.Johnson C. et al.Magnetoencephalography with a two-color pump-probe, fiber-coupled atomic magnetometer.Appl. Phys. Lett. 2010; 97243703Crossref Scopus (132) Google Scholar]. By 2013, the same group had demonstrated a multisensor OPM array [7.Johnson C.N. et al.Multi-sensor magnetoencephalography with atomic magnetometers.Phys. Med. Biol. 2013; 58: 6065-6077Crossref PubMed Scopus (82) Google Scholar]. Other demonstrations included the use of a chip-scale OPM to measure evoked and oscillatory neuromagnetic effects [11.Sander T.H. et al.Magnetoencephalography with a chip-scale atomic magnetometer.Biomed. Opt. Express. 2012; 3: 981-990Crossref PubMed Scopus (206) Google Scholar] and measurement of the modulation of occipital alpha oscillations by opening and closing the eyes [31.Kamada K. et al.Human magnetoencephalogram measurements using newly developed compact module of high-sensitivity atomic magnetometer.Jpn. J. Appl. Phys. 2015; 54026601Crossref Scopus (52) Google Scholar]. A significant step forward came with the introduction of commercial OPMs, by QuSpin Inc. in 2016, allowing the neuroimaging community to begin to build OPM-MEG systems. This drove further progress and soon multiple groups had begun to demonstrate the promise of robust and reliable microfabricated OPMs for MEG measurement (e.g., [14.Boto E. et al.A new generation of magnetoencephalography: room temperature measurements using optically-pumped magnetometers.NeuroImage. 2017; 149: 404-414Crossref PubMed Scopus (198) Google Scholar,32.An K. et al.Detection of the 40-Hz auditory steady-state response with optically pumped magnetometers.BioRxiv. 2021; (Published online October 3, 2021)https://doi.org/10.1101/2021.10.01.462598Google Scholar, 33.Iivanainen J. et al.On-scalp MEG system utilizing an actively shielded array of optically-pumped magnetometers.NeuroImage. 2019; 194: 244-258Crossref PubMed Scopus (76) Google Scholar, 34.Iivanainen J. et al.Potential of on-scalp MEG: robust detection of human visual gamma-band responses.Hum. Brain Mapp. 2019; 41: 150-161Crossref PubMed Scopus (26) Google Scholar]).A major limitation of conventional MEG (and fMRI) is limited tolerance to participant movement. Because the sensor array is fixed, any motion of the participant relative to the array causes changes in signal amplitude and SNR (as brain regions get closer to, or further from, the sensors) and spatial blurring of the field topography. Consequently, MEG signals from conventional scanners become distorted, in time and space. Recent years have seen the development of algorithms to measure (in real time) and correct (in post-processing) such artefacts (e.g., [35.Messaritaki E. et al.Assessment and elimination of the effects of head movement on MEG resting-state measures of oscillatory brain activity.NeuroImage. 2017; 159: 302-324Crossref PubMed Scopus (10) Google Scholar, 36.Stolk A. et al.Online and offline tools for head movement compensation in MEG.NeuroImage. 2013; 68: 39-48Crossref PubMed Scopus (124) Google Scholar, 37.Taulu S. Simola J. Spatiotemporal signal space separation method for rejecting nearby interference in MEG measurements.Phys. Med. Biol. 2006; 51: 1759Crossref PubMed Scopus (868) Google Scholar, 38.Taulu S. et al.Applications of the signal space separation method.IEEE Trans. Signal Process. 2005; 53: 3359-3372Crossref Scopus (363) Google Scholar]) and the importance of these methods, particularly in paediatric imaging, is well recognised [39.Chen Y.-H. et al.Magnetoencephalography and the infant brain.NeuroImage. 2019; 189: 445-458Crossref PubMed Scopus (24) Google Scholar, 40.Wehner D.T. et al.Head movements of children in MEG: quantification, effects on source estimation, and compensation.NeuroImage. 2008; 40: 541-550Crossref PubMed Scopus (53) Google Scholar, 41.Nenonen J. et al.Validation of head movement correction and spatiotemporal signal space separation in magnetoencephalography.Clin. Neurophysiol. 2012; 123: 2180-2191Crossref PubMed Scopus (40) Google Scholar]; a simulation study [42.Larson E. Taulu S. The importance of properly compensating for head movements during MEG acquisition across different age groups.Brain Topogr. 2017; 30: 172-181Crossref PubMed Scopus (15) Google Scholar] has shown that, whilst head movement results in a significant degradation in spatial accuracy, with appropriate compensation, accuracy can be restored to premovement levels even in the presence of small (e.g., 2–3 cm) movements. These techniques can allow high-fidelity MEG acquisition in infants and patient groups who find it hard to remain still. However, other studies have suggested complications due to changes in SNR as sources move relative to the sensors, potentially placing upper limits on the magnitude of movement that can be compensated [43.Medvedovsky M. et al.Artifact and head movement compensation in MEG.Neurol Neurophysiol. Neurosci. 2007; 4: 1-10Google Scholar]. Most importantly, even with movement compensation, successful MEG measurement relies on the subject's head remaining inside the helmet and this places a hard limit on the allowable head movement. For adults, movement is physically restricted (i.e., a movement of more than a few centimetres would cause the subject to hit their head on the helmet), which limits the ability to perform naturalistic tasks. In infants, the requirement to remain within the helmet can be met with training [44.Rapaport H. et al.Studying brain function in children using magnetoencephalography.J. Vis. Exp. 2019; (Published online April 8, 2019)https://doi.org/10.3791/58909-vCrossref PubMed Scopus (4) Google Scholar], but many infants still find the unnatural environment difficult to tolerate and again this restricts experimental paradigms. This has been one of the major limitations of MEG/fMRI compared with EEG or functional near infra-red spectroscopy (fNIRS) [45.Pinti P. et al.The present and future use of functional near-infrared spectroscopy (fNIRS) for cognitive neuroscience.Ann. N. Y. Acad. Sci. 2020; 1464: 5-29Crossref PubMed Scopus (250) Google Scholar], both of which involve wearable instrumentation where movement is not curtailed by a static helmet (or for fMRI, an enclosed scanner).By contrast, the lightweight nature of OPMs means that sensors, mounted in a suitable helmet (Box 2), can move with the head. Consequently, the MEG scanner can become a wearable device, allowing (in principle) any
Referência(s)