Fabrication of a spherical inclusion phantom for validation of magnetic resonance-based magnetic susceptibility imaging.
Jun-Ho KimJung-Hyun KimSo-Hee LeeJinhyoung ParkSeung-Kyun LeePublished in: PloS one (2019)
Fabrication of a spherical multi-compartment MRI phantom is demonstrated that can be used to validate magnetic resonance (MR)-based susceptibility imaging reconstruction. The phantom consists of a 10 cm diameter gelatin sphere that encloses multiple smaller gelatin spheres doped with different concentrations of paramagnetic contrast agents. Compared to previous multi-compartment phantoms with cylindrical geometry, the phantom provides the following benefits: (1) no compartmental barrier materials are used that can introduce signal voids and spurious phase; (2) compartmental geometry is reproducible; (3) spherical susceptibility boundaries possess a ground-truth analytical phase solution for easy experimental validation; (4) spherical geometry of the overall phantom eliminates background phase due to air-phantom boundary in any scan orientation. The susceptibility of individual compartments can be controlled independently by doping. During fabrication, formalin cross-linking and water-proof surface coating effectively blocked water diffusion between the compartments to preserve the phantom's integrity. The spherical shapes were realized by molding the inner gel compartments in acrylic spherical shells, 3 cm in diameter, and constructing the whole phantom inside a larger acrylic shell. From gradient echo images obtained at 3T, we verified that the phantom produced phase images in agreement with the theoretical prediction. Factors that limit the agreement include: air bubbles trapped at the gel interfaces, imperfect magnet shimming, and the susceptibility of external materials such as the phantom support hardware. The phantom images were used to validate publicly available codes for quantitative susceptibility mapping. We believe that the proposed phantom can provide a useful testbed for validation of MR phase imaging and MR-based magnetic susceptibility reconstruction.