Noise-free Magnetoencephalograpy (MEG) recordings of Brain Function


Magnetoencephalography (MEG) is a completely noninvasive method of ‘imaging’ human brain activity by measuring the magnetic fields produced by neuronal activity. Perhaps the greatest impediment to acquiring high-quality MEG recordings is the ubiquitous ambient magnetic field noise produced by elevators, trains, cars, and even room lights. A team of scientists led by Dr. Robert Kraus, Jr., at Los Alamos National Laboratory with DOE support has recently combined two techniques that enables to completely eliminate magnetic noise (which can be up to 1,000,000 times larger than the signals from the brain) from MEG recordings. The team invented the idea of using a superconducting helmet (called a SIS) to surround the array of superconducting quantum interference device sensors (or SQUIDs, the most sensitive magnetic field sensors known) that measure the MEG signal. Because magnetic fields cannot penetrate the SIS, it reduces the background fields at the MEG sensors by up to 1000 times. The team recently added ‘reference sensors’ located on the outside of the superconducting helmet to measure ambient magnetic fields. While the reference sensors are perfectly shielded from all brain sources, they are in very close proximity to the MEG sensors. This arrangement enables very accurate estimation and subtraction of the ambient field noise contribution to the MEG sensors using an adaptive computer algorithm. They found this technique reduced the ambient magnetic noise by more than 1,000,000 times! The residual noise for most MEG SQUID sensors is the noise intrinsic to the sensors and electronics allowing them to measure human brain activity better than ever before. This work has been submitted to the journal: Physics in Medicine and Biology.