Investigating methods for temperature reconstruction based on simulated data

Magnetic particle imaging (MPI) leverages the nonlinear response of magnetic particles for imaging. Since the response of the particles is dependent on environmental parameters like temperature or viscosity it can reconstruct images of these properties. It has been shown that MPI is a promising tool to monitor temperatures using a reconstruction approach called multi-color MPI, which enables temperature imaging in applications such as magnetic particle hyperthermia. This work investigates different approaches to reconstruct temperature differences as well as different reconstruction schemes for multi-color MPI. The performed reconstructions are based on simulated 2D system matrices providing a reliable baseline for comparisons. The simulated system matrices are obtained by simulations of the magnetic particles in different magnetic offset fields corresponding to the spatial positions in a gradient field.