An Application-Optimized Array Compressed Parallel Transmit RF System for Ultra-high Field Magnetic Resonance Imaging

Abstract

Ultra-high field (UHF; 7 Tesla and higher) MRI provides enhanced sensitivity, contrast, and spatial resolution for neuroimaging exams than lower-field scanners. However, at the high Larmor frequencies of UHF scanners, transmit fields become spatially inhomogeneous, leading to spatially varying signal and contrast. Radiofrequency (RF) shimming using a parallel transmit (pTx) coil array can homogenize transmit fields but is limited by the number of transmit channels available on conventional UHF scanners (typically 8). This project aims to address this limitation through the development of a process and hardware solution to optimally drive a large number of RF coils with a small number of transmit channels, called array compressed parallel transmission (acpTx). We first present an open-source design framework for compact, low-loss microstrip unequal power splitters, which are the building blocks for an array compression network. We then present a 30-element transmit coil which is the densest head transmit array built to date, and which provides the degrees of freedom needed to tailor the compressed transmit channels' RF fields. Finally, we combine these developments to achieve the first integrated array compressed parallel transmit system which is optimized for 2D multislice RF-shimmed brain imaging. More broadly, this work represents the first time that the MRI spin physics of a specific imaging acquisition has been directly leveraged to design an optimal RF transmit coil for that imaging sequence.