International Journal on Magnetic Particle Imaging IJMPI
Vol 9 No 1 Suppl 1 (2023): Int J Mag Part Imag
https://doi.org/10.18416/IJMPI.2023.2303077

Proceedings Articles

Deconvolution of direct reconstructions for MPI using Convolutional Neural Network

Main Article Content

Mathias Eulers  (Universität zu Lübeck), Christine Droigk  , Marco Maass  , Alfred Mertins  

Abstract

Recently, an approach was presented that allows a direct and fast image reconstruction without the use of a systemmatrix for Lissajous trajectories of the so called field free point. The method is based on weighting frequencycomponents of the measured voltage signals and additional factors with Chebychev polynomials of the secondkind, resulting in reconstructions of the convolved spatial distribution of magnetic nanoparticles. In order toobtain meaningful images, these reconstructions have to be deconvolved afterwards. For this purpose, differentmethods have already been proposed. In this work, a U-shaped neural network is used for the deconvolution. Thenetwork was trained and tested on simulated data of blood vessel like structures. The proposed model outperformsconventional methods and improves the image quality of the reconstructions.

Article Details

Most read articles by the same author(s)

1 2 > >>