Vortrag: "Machine Learning and Model-Based Reconstruction in Magnetic Particle Imaging"

Magnetic Particle Imaging (MPI) is a relatively new imaging modality that is currently in the preclinical phase. Potential fields of application are angiography, in which magnetic nanoparticles are injected into the bloodstream to visualize the vascular tree, or cancer detection, in which the nanoparticles are absorbed much more quickly by cancerous than by normal tissue. In MPI, the measured signal consists of a voltage measured in receive coils induced by the magnetization response of the magnetic particles experiencing a magnetic field. The reconstruction of the particle distribution from this signal constitutes an ill-posed inverse problem. In this talk, we present our results on model based reconstruction, and on the use of machine learning in this context.