Robust Magnetic Field Homogeneity Enhancement via Local Passive Shimming in a Superconducting Magnet for Animal MRI

In ultrahigh-field magnetic resonance imaging (MRI) systems, achieving subppm static magnetic field homogeneity is critical. However, engineering tolerances and assembly errors often introduce complex field inhomogeneities that global shimming alone cannot always fully correct. In this study, a novel local passive shimming method is developed for small-bore MRI superconducting magnets. By establishing a linear programming model for local shimming, the approach enables region-specific correction of high-order field variations. Numerical simulations and experimental verification on a 7 T/300-mm superconducting magnet demonstrate significant improvements in homogeneity, reducing root mean square error (RMSE) and peak-to-peak variation across the diameter of spherical volume (DSV). The optimized shim configuration achieved a peak-to-peak inhomogeneity of 4.23 ppm within the DSV, indicating substantial enhancement over baseline performance. These results indicate that the proposed technique serves as a practical and efficient complement to global shimming strategies, offering a robust solution for field correction in small-bore MRI systems.