MedGAN: Medical image translation using GANs

Karim Armanious, Chenming Jiang, Marc Fischer, Thomas Küstner, Tobias Hepp, Konstantin Nikolaou, Sergios Gatidis, Bin Yang

Abstract: Highlights•MedGAN as a new end-to-end framework for medical image translation.•Combination of cGAN with non-adversarial losses and a new generator architecture.•Application on several medical tasks with no modifications to the hyperparameters.•MedGAN outperforms other approaches in qualitative and quantitative comparisons.•Perceptual evaluation was performed by five experienced radiologists.