Experimental Study of a Room-Temperature Shimming Technology Employing Genetic Algorithm for NMR/MRI Superconducting Magnets

Abstract

We present a recently developed shimming method that uses a system of room-temperature (RT) orthogonal shimming coils to improve the spatial field homogeneity of high-resolution nuclear magnetic resonance (NMR) and magnetic resonance imaging (MRI) superconducting magnets. Conventional RT shimming methods require recursive routines of field mapping and RT shim coil current adjustments, which lead to a less effective and time-consuming process. In this paper, we introduce a new RT shimming method that utilizes a process of gradual and automatic search to find an optimal operating point with a genetic algorithm (GA), a type of optimization technique. The proposed algorithm is capable of determining the optimal set of currents for a system of shimming coils to create the best spatial field homogeneity by minimizing the NMR spectral linewidth. We used a commercial 600-MHz NMR magnet with a full NMR spectral linewidth at half maximum (FWHM) of 2.2 ppm, as determined by an NMR test sample contained in a cylindrical glass tube with a diameter of 5 mm to demonstrate that the shimming method can improve the FWHM to 0.15 ppm. The results demonstrate the strong potential of the proposed RT shimming methodology for the auto-shimming process of NMR/MRI applications.