Editorial: Magnetic Neurophysiology: The Cutting Edge of Real Time Neurodiagnostic Technology

Pegah Afra^1*Shigenori Kawabata²Glenn D.R. Watson³Timothy Roberts^4*

• ¹University of Massachusetts Medical School, Worcester, United States
• ²Institute of Science Tokyo, Tokyo, Japan
• ³Duke University, Durham, United States
• ⁴The Children's Hospital of Philadelphia, Philadelphia, United States

Magnetoneurophysiology, the study of magnetic fields generated by neural activity, has experienced a significant evolution in recent years, driven by advances in sensor technology and a renewed focus on clinical applications. The field is grounded in fundamental biophysics, namely the intra-axonal currents of action potentials that produce measurable magnetic fields. While these principles have been long understood, emerging technologies such as optically pumped magnetometers (OPMs) and improved superconducting quantum interference devices (SQUIDs) have transformed both the precision and accessibility of magnetographic recordings. This editorial brings together a series of reviews and research contributions to our research topic that collectively chart the trajectory from foundational principles to cutting-edge innovations in magnetoencephalography (MEG), magnetospinography (MSG), and magnetoneurography (MNG), and their combination. These developments not only enhance technical performance, but also open the door to novel, more naturalistic paradigms for studying brain, spinal cord, and peripheral nerve activity that can shape the future of clinical neurophysiology. The intra-axonal flow of the propagating action potential generates a magnetic field that is unaffected by volume conducting properties of the surrounding tissue, although signal decay with distance remains an issue (1). A review by Adachi and Kawabata (2) provides a clear explanation of the fundamentals of magnetoneurophysiology. Specifically, they highlight the mechanisms that underlies magnetic fields traveling the spinal cord and peripheral nerves: Transmembrane inward and outward currents generated during action potentials, along with intracellular leading and trailing currents and extracellular volume currents. These relationships are clearly described and illustrated in their article's Figure 2. Although these concepts have been clearly studied in clinical neurophysiology, they have rarely been depicted in one clear, easy to comprehend figure. The different sensor-level technologies are discussed in detail in our research topic series. Adachi have led to significant differences between OPM sensors even of the same vapor type. Moreover, fundamental considerations like "noise level" differ between systems with, for example, He-OPMs typically characterized by noise levels, two to three times greater than their Rb counterparts. Key parameters thus have to be assessed critically in designing OPM systems for MEG. These include differences in operating bandwidth (for the range of signals MEG might detect -typical clinical applications consider activity in the 0.1-100Hz range), signal to noise ratio (dependent on the noise level, with is typically <10fT/ÖHz for conventional SQUID systems and can approach this for Rb-OPM (~10-15fT/ÖHz), but tends to be 2-3x more challenging for He-OPM, ~30-50fT/ÖHz) (3)(4)(5)(6)(7)(8). Further, the dynamic range of the detectors, and their associated electronics, influences their tolerances of different ambient magnetic fields (and thus the degree to which the arrays can "move", with the participant, through spatially varying fields). These differences have implications for clinical applications: SQUID sensors require a rigid helmet, inside which the sensors rest in a cryogenic liquid-He dewar (resulting in increased distance of sensors from source, with concomitant loss of sensitivity). On the other hand, heated Rb-based OPMs require thermal insulation and/or airflow, while the He-based systems, do not have any reasons for distancing, allowing them to be placed close to the scalp. In clinical practice, with the thin insulation, Rb-OPMs can also be considered "scalp-mounted". The main advantage of SQUID sensors is their bandwidth (DC-up to 40 kHz) compared to Rb-OPMs (DC-200 Hz) and He-OPMs (DC-2 kHz), although given the range of signals detected from the human brain, this advantage may currently be considered of little practical impact. Optically pumped magnetometer detector systems also are characterized by a wide variation in dynamic operating range from ~5nT to up to 300nT, often dependent on their operation in closed loop (vs. open loop, with lower dynamic range); however, other factors such as feedback electronics influence achievable dynamic range. While the lower limit of this might preclude movement through spatially-varying magnetic fields, as shown by Roberts (7) some commercial systems indeed have sufficient dynamic range (~150nT) to tolerate head movement and allow effective recording from the moving head. Nonetheless, similar to SQUIDs, OPM sensors currently have to operate in a magnetically shielded room (MSR), in which spatial variations of magnetic field can typically be maintained less than 100nT, which can thus tolerate the OPM array (and thus the participant) moving to different positions in the room (for OPM sensors with appropriate dynamic range >~100nT). Cross axis projection errors (CAPE) are discussed by Spedden et al (9), and how the large background magnetic noise can introduce gain and orientation errors in Rb-based OPM output signal for some systems requiring field nulling systems (closed loop mode of operation, where internal coils null the field), although technical differences do exist even within different Rb-OPM systems, with some systems demonstrating robustness to these concerns. Bonnet et al. (3) explains that the He-based OPMs due to their large dynamic range result in less saturation and therefore require no additional field nulling systems, although as shown by Roberts et al. (7), Rb-OPM systems with sufficient dynamic range can also operate effectively without additional external field nulling. In the cryo-space, Roberts et al. (7) reviews the traditional SQUID-based MEG systems, including the requirement for a rigid helmet to host SQUID-based sensors in a cryo-bath. Practical constraints include a need to maintain vacuum and thermal insulation within the dewar, which translates into physical inflexibility of the fixed sensory array, and a need for dedicated infant, child and adult MEG dewar sizes, each adding cost to the MEG laboratory alongside to the cost of maintaining vacuum and weekly helium refills to maintain the cryo-bath.Adachi and Kawabata (2), discuss the latest SQUID-based MSG/MNG system that is suitable for recording magnetic fields from spinal cord, plexi and peripheral nerves, employing a new kind of machine that has never been available before. Cryogenic MSG/MNG developed by the Kazanawa