2019/08/01 by Tycho F. A. van der Ouderaa, van der Ouderaa, Tycho F. A., Daniel E. Worrall +3 · 2 citations
Computer Science · Earth and Planetary Sciences · Engineering · Mathematics · Neuroscience · #Adaptation (eye) #Advanced Image Processing Techniques #Artificial intelligence #Artificial neural network #Biology #Computer science #Computer vision #Deep learning #Domain (mathematical analysis) #Domain adaptation #FOS: Electrical engineering #Field (mathematics) #Image (mathematics) #Image and Signal Denoising Methods #Image and Video Processing (eess.IV) #Image resolution #Image translation #Invertible matrix #Mathematics #Neuroscience #Pattern recognition (psychology) #Property (philosophy) #Resolution (logic) #Seismic Imaging and Inversion Techniques #Task (project management) #Translation (biology) #eess.IV #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1908.00295
published in arXiv (Cornell University) (Cornell University) · MIDL 2019 [arXiv:1907.08612]
arxiv created 2019/08/01 · openalex publication_date 2019/08/01 · arxiv updated 2019/08/02 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
Recently, paired (e.g. Pix2pix) and unpaired (e.g. CycleGAN) image-to-image translation methods have shown effective in medical imaging tasks. In practice, however, it can be difficult to apply these deep models on medical data volumes, such as from MR and CT scans, since such data volumes tend to be 3-dimensional and of a high spatial resolution pushing the limits of the memory constraints of GPU hardware that is typically used to train these models. Recent studies in the field of invertible neural networks have shown that reversible neural networks do not require to store intermediate activations for training. We use the RevGAN model that makes use of this property to perform memory-efficient partially-reversible image-to-image translation. We demonstrate this by performing a 3D super-resolution and a 3D domain-adaptation task on chest CT data volumes.