Improved image reconstruction in magnetic particle imaging using structural a priori information

Magnetic particle imaging is a tracer-based imaging modality developed to detect the concentration of superparamagnetic iron oxide nanoparticles. The capability for imaging is due to the high sensitivity to the nanoparticle’s nonlinear response to the applied magnetic field. This modality relies on the spatial distribution of the tracer material which makes it suitable for applications such as imaging blood flow or tracking medical instruments without the need of harmful radiation. Magnetic particle imaging benefits from a high temporal resolution, but it also suffers from missing background information, e.g., from biological tissue. Commonly the lack of information is remedied by magnetic resonance imaging. Image reconstructions from both modalities are computed independently and aligned subsequently to allow inferences. We use the additional information commonly provided by magnetic resonance imaging to improve the reconstruction in magnetic particle imaging. For this purpose, a Tikhonov-type functional is equipped with a structural prior where the additional information is incorporated. By minimizing this functional, we obtain improved reconstructions of the concentration of nanoparticles which is illustrated in numerical simulations.