International Journal on Magnetic Particle Imaging IJMPI
Vol. 9 No. 1 Suppl 1 (2023): Int J Mag Part Imag
GAN-based deblurring of reconstructed images for MPI
Main Article Content
Abstract
Reconstructed images may suffer from blurring and reconstruction artefacts in Magnetic Particle Imaging. In this work, a generative adversarial network is used for deblurring reconstructed images in a post-processing step. The network is trained using eigen-reconstructions of a system matrix and evaluated on synthesized phantoms.