F8 - Projection Domain Metal Artifact Reduction in Computed Tomography using Conditional Generative Adversarial Networks
Nele Blum, Thorsten Buzug, Maik Stille
High-density objects in the field of view, still remain one of the major challenges in CTimage reconstruction. They cause artifacts in the image, which degrade the quality andthe diagnostic value of the image. Standard approaches for metal artifact reduction areoften unable to correct these artifacts sufficiently or introduce new artifacts. In this work,a new deep learning approach for the reduction of metal artifacts in CT images is proposedusing a Generative Adversarial Network. A generator network is applied directly to theprojection data corrupted by the metal objects to learn the corrected data. In addition, asecond network, the discriminator, is used to evaluate the quality of the learned data. Theresults of the trained generator network show that most of the data could be reasonablyreplaced by the network, reducing the artifacts in the reconstructed image.
Thursday 8th July
F1-9 (short): Imaging: Reconstruction and Clinical Data - 13:45 - 14:30 (UTC+2)