Information content with low- vs. high-Tc SQUID arrays in MEG recordings: The case for high-Tc SQUID-based MEG
Artikel i vetenskaplig tidskrift, 2014

Background: Magnetoencephalography (MEG) is a method of studying brain activity via recordings of the magnetic field generated by neural activity. Modern MEG systems employ an array of low critical-temperature superconducting quantum interference devices (low-Tc SQUIDs) that surround the head. The geometric distribution of these arrays is optimized by maximizing the information content available to the system in brain activity recordings according to Shannon's theory of noisy channel capacity. New method: Herein, we present a theoretical comparison of the performance of low- and high-Tc SQUID-based multichannel systems in recordings of brain activity. Results: We find a high-Tc SQUID magnetometer-based multichannel system is capable of extracting at least 40% more information than an equivalent low-Tc SQUID system. The results suggest more information can be extracted from high-Tc SQUID MEG recordings (despite higher sensor noise levels than their low-Tc counterparts) because of the closer proximity to neural sources in the brain. Comparison with existing methods: We have duplicated previous results in terms of total information of multichannel low-Tc SQUID arrays for MEG. High-Tc SQUID technology theoretically outperforms its conventional low-Tc counterpart in MEG recordings. Conclusions: A full-head high-Tc SQUID-based MEG system's potential for extraction of more information about neural activity can be used to, e.g., develop better diagnostic and monitoring techniques for brain disease and enhance our understanding of the working human brain. © 2013 Elsevier B.V.

priority journal

SQUID-sensor arrays

High-Tc SQUID

magnetoencephalography

magnetic field

cerebrospinal fluid

analyzer

signal noise ratio

superconducting quantum interference device

article

electroencephalogram

Total information

Channel capacity

sensor

noise

process optimization

MEG

white noise

Neuroimaging

recording

Författare

Justin Schneiderman

Göteborgs universitet

Journal of Neuroscience Methods

0165-0270 (ISSN)

Vol. 222 42-46

Ämneskategorier

Neurovetenskaper

DOI

10.1016/j.jneumeth.2013.10.007