University of Heidelberg
Faculty of Medicine Mannheim
University Hospital Mannheim
CKM receives DFG funding for Project "Prädiktion von Therapieansprechen und Outcome beim lokal fortgeschrittenen Rektum-Karzinom mittels Radiomics und Deep Learning: eine beispielhafte Anwendung für eine allgemein verwendbare, Deep Learning basierte Prozessierungs-Pipeline für die Bild-Klassifikation" read more.
Alena-Kathrin Schnurr presented on our research on AI in medical imaging @ Research Plus forum
read more
Dr. Mathias Davids wins the I. I. Rabi Award of the International Society for Magnetic Resonance in Medicine (ISMRM),
read more, article in local newspaper (Mannheimer Morgen)

If you have questions concerning a specific publication please use this form with subject 'information about publications' and giving the full citation in the message body.

Home > Publications > Abstract >

Synthesis of {\{}CT{\}} Images Using Cycle{\{}GAN{\}}s: Enhancement of Anatomical Accuracy

D. Bauer, A. Schnurr, T. Russ, S. Goerttler, L. Schad, F. Zoellner and K. Chung

International Conference on Medical Imaging with Deep Learning - Extended Abstract Track,

Deep learning in medical imaging is often limited by the availability of training data with sufficient quality. The synthesis of image data offers a solution to this data shortage. Here, we use the CycleGAN network architecture to synthesize axial CT slices based on anthropomorphic body phantoms. We investigate the influence of an identity loss and a gradient difference loss function on the image quality of the synthesized data. We evaluate the synthesized images with respect to anatomical accuracy and realistic CT noise properties. The additional loss functions improved the preservation of edges and anatomical structures compared to the original CycleGAN loss, without deteriorating the noise quality of the synthetic image.

Contact: Prof. Dr. Frank Zöllner last modified: 01.07.2020
to top of page