Efficient Iterative Reconstruction for an MPI Equilibrium Model with Anisotropy

Image reconstruction in Magnetic Particle Imaging (MPI) typically requires a system matrix, obtained through a time-consuming calibration process. To bypass this, various model-based approaches have been explored. Recent work demonstrated successful reconstruction by adapting a Chebyshev approach with Tikhonov-regularized least squares (LS) under an equilibrium model with anisotropy. In this study, we introduce an efficient evaluation of the forward and adjoint operators for the anisotropy model, enabling the use of iterative solvers and alternative regularization methods for image reconstruction.