A 90-channel triaxial magnetoencephalography system using optically pumped magnetometers

Magnetoencephalography (MEG) measures the small magnetic fields generated by current flow in neural networks, providing a noninvasive metric of brain function. MEG is well established as a powerful neuroscientific and clinical tool. However, current instrumentation is hampered by cumbersome cryogenic field-sensing technologies. In contrast, MEG using optically pumped magnetometers (OPM-MEG) employs small, lightweight, noncryogenic sensors that provide data with higher sensitivity and spatial resolution, a natural scanning environment (including participant movement), and adaptability to any age. However, OPM-MEG is new and the optimum way to design a system is unknown. Here, we construct a novel, 90-channel triaxial OPM-MEG system and use it to map motor function during a naturalistic handwriting task. Results show that high-precision magnetic field control reduced background fields to similar to 200 pT, enabling free participant movement. Our triaxial array offered twice the total measured signal and better interference rejection compared to a conventional (single-axis) design. We mapped neural oscillatory activity to the sensorimotor network, demonstrating significant differences in motor network activity and connectivity for left-handed versus right-handed handwriting. Repeatability across scans showed that we can map electrophysiological activity with an accuracy similar to 4 mm. Overall, our study introduces a novel triaxial OPM-MEG design and confirms its potential for high-performance functional neuroimaging.

Automated summary

Magnetoencephalography (MEG) is a noninvasive technique that measures the magnetic fields generated by neural networks. Conventional MEG instruments use cryogenic field-sensing technologies, while a new approach called optically pumped magnetometers MEG (OPM-MEG) uses lightweight sensors with higher sensitivity and resolution. This study introduces a novel triaxial OPM-MEG system and demonstrates its potential for high-performance functional neuroimaging by mapping motor function during a handwriting task.