Open-source software and hardware tools for local B0 field control

点击翻译按钮获取中文摘要

Project Summary/Abstract Multi-coil (MC) shim arrays have emerged as a promising and flexible tool for improving MRI image quality. Arrays of small, independently-driven loops placed close to the body provide an efficient way to generate rapidly-switchable magnetic field offsets (ΔB0) that can be shaped to provide useful field profiles inside the body. MC arrays were originally proposed for dynamically-switchable, high spatial order “B0 shimming” in the body to null subject-specific perturbations of the static background B0 field. The improved shimming reduces geometric distortion in echo planar imaging (widely used for functional and diffusion MRI) and line broadening in MR spectroscopy. However, in the past few years, a surge of new uses for MC arrays have been proposed, including supplementary spatial encoding, improved lipid suppression, zoomed imaging, and reduced flip angle (B1+) inhomogeneity. This diverse and growing set of methods – which we classify as local field control – exploit two core features of MC arrays: (1) the ability of non-orthogonal ΔB0 basis sets to generate field profiles that can not be created with linear gradients; and (2) the ability to rapidly update shim currents without causing artifacts. Unfortunately, MC local field control research has been slow to spread beyond a small handful of sites due to limited availability of instrumentation as well as control software. Commercial shim amplifiers with dynamic switching capability are rare, and those that do exist are cost-prohibitive for most applications (>$1,000/channel). At the same time, there is no readily-available software for controlling shim amplifiers and interfacing with the scanner host computer. Moreover, there is a lack of software tools using convex optimization to efficiently solve for shim current amplitudes for tailored local field control. We will break down these barriers to entry by developing an open-source resource called AFFECT (Automated Flexible Field Encoding and Control Toolkit). We will refine and disseminate our previously-validated low-cost ($100- 150/channel), low-voltage shim amplifier that is scalable up to 64-channels. We will also upgrade and package a graphical-user-interface (GUI) used to process B0 field maps and compute optimal shim currents. The GUI will be made modular to allow users to plug in their own custom shim optimization tools. Finally, we will create a seamless interface between the GUI and the scanner host computer to improve workflow. More than 10 research groups have already contacted us asking to use our open-source shim amplifiers. However, further work is required to prepare both the hardware and software for dissemination. The goal of this project is to translate our prototypes into robust, user-extensible tools that are packaged and documented. To expedite dissemination, we will provide up to 10 research groups with 32-channel amplifier setups free of charge. Users will also be free to download schematics and fabricate the circuit boards on their own.