Integrated Parallel Reception, Excitation, and Shimming (iPRES) Head Coils: Application to Functional MRI and Development of a Battery Powered System

In MRI, magnetic susceptibility differences at air/tissue interfaces result in main magnetic field (B0) inhomogeneities, which in turn cause artifacts such as distortions and signal loss, especially in gradient-echo echo-planar imaging (EPI). Because of the air cavities in the sinuses and ear canals, these artifacts are most prominent in the inferior frontal and temporal brain regions, and hinder our ability to accurately assess brain structure and function. Existing shimming techniques can correct for these artifacts by applying additional magnetic fields that compensate for the B0 inhomogeneities. However, passive shimming has limited flexibility and patient comfort, whereas spherical harmonic shimming cannot shim localized B0 inhomogeneities.

Integrated parallel reception, excitation, and shimming (iPRES) is a novel technology that allows radio-frequency and direct currents (DC) to flow in the same coil simultaneously, enabling image acquisition and localized B0 shimming, respectively. By using a single coil array placed close to the subject, iPRES saves valuable space in the magnet bore and maximizes both the signal-to-noise ratio and shimming performance. Two advancements to this technology are proposed in this thesis: (1) a method to effectively recover signal loss in functional MRI (fMRI) and (2) a plug-and-play battery- powered iPRES head coil array.

Previously, iPRES has been used to correct for distortions in EPI images. In this work, we propose to use it to recover signal loss in fMRI. Three types of approaches were simulated: modifications of the coil geometry, adjustments to the volume shimmed, and redefinition of the cost function used in the shim optimization. The simulation results from the modified cost function were the most promising and showed that there was a tradeoff between reduction in B0 inhomogeneity and signal loss, which could be adjusted by using a weight.

The shimming performance was further assessed in in vivo fMRI experiments by acquiring B0 maps and EPI images during a breathholding task, before and after shimming with the optimal weight. The results confirmed that the proposed method could effectively recover signal loss in fMRI, while also reducing the B0 inhomogeneity.

Existing iPRES coil arrays use a multi-channel DC power supply, which allows the DC currents to be optimized to shim individual subjects and/or slices, giving the best shimming performance. However, this implementation requires many cables and filters between the coil array and the machine room. Additionally, subject-specific shimming involves the acquisition of a B0 map and a DC current optimization for each subject, which requires additional time and technical expertise, and may not be practical for clinical applications.

In this work, an MR-compatible battery pack was developed to power an iPRES head coil array, eliminating the external DC power supply, cables, and filters, and reducing the cost and complexity of the system. The battery pack was designed to deliver fixed DC currents to a subset of iPRES coil elements, which were optimized in advance to shim an average subject’s brain, thus eliminating the subject-specific shim optimization.

The system was validated through bench-top measurements and MRI experiments to verify that the shim currents were accurate and remained stable within the operating time. B0 maps and EPI images acquired in vivo before and after shimming with the battery- powered iPRES head coil array showed that it could reduce the B0 inhomogeneity and geometric distortion. While it does not offer the flexibility to shim individual subjects or slices, this stand-alone, plug-and-play system is expected to enable a wider adoption of iPRES in clinical applications.