Our team (Thomas Germer, Jan Robine, Sebastian Konietzny, Stefan Harmeling, and Tobias Uelwer) won the Helsinki Tomography Challenge 2022 (HTC 2022) presented by the Finnish Inverse Problems Society (FIPS).
Given the small amount of data provided, we generate a large number of synthetic images by placing various shapes on top of a disk. We then compute sinograms from those images using the forward operator from the Helsinki Tomography Toolbox and Astra Toolbox. Lastly, we train a convolutional neural network to directly predict the images from the subsampled sinograms.
About the HTC 2022: www.fips.fi/HTCabout.ph
Challenge Results: www.fips.fi/HTCresults.php