Demonstration of full tensor current density imaging using ultra-low field MRI

TL;DR

This paper demonstrates a novel ultra-low field MRI technique using SQUID sensors to directly image the full tensor of current density in phantoms and provides insights for in vivo applications, potentially advancing neuroimaging.

Contribution

It introduces a full tensor current density imaging method at ultra-low magnetic fields using SQUID-based MRI without subject rotation, enabling new neuroimaging possibilities.

Findings

Successful 3D current density imaging on phantoms

Achieved detection of currents around 2 mA with specific current densities

Identified necessary improvements for in vivo signal-to-noise ratio

Abstract

Direct imaging of impressed dc currents inside the head can provide valuable conductivity information, possibly improving electro-magnetic neuroimaging. Ultra-low field magnetic resonance imaging (ULF MRI) at $\mu$T Larmor fields can be utilized for current density imaging (CDI). Here, a measurable impact of the magnetic field $B_{J}$, generated by the impressed current density $J$, on the MR signal is probed using specialized sequences. In contrast to high-field MRI, the full tensor of $B_{J}$ can be derived without rotation of the subject in the scanner, due to a larger flexibility in the sequence design. We present an ULF MRI setup based on a superconducting quantum nterference device (SQUID), which is operating at a noise level of 380 aT Hz$^{-1/2}$ and capable of switching all imaging fields within a pulse sequence. Thereby, the system enables zero-field encoding, where the full tensor of $B_{J}$ is probed in the absence of other magnetic fields. 3D CDI is demonstrated on phantoms with different geometries carrying currents of approximately 2 mA corresponding to current densities between 0.45 and 8 A/m$^{2}$. By comparison to an in vivo acquired head image, we provide insights to necessary improvements in signal-to-noise ratio.