MedGAN: Medical Image Translation using GANs

By Karim Armanious, Chenming Yang, Marc Fischer, Thomas Küstner, Konstantin Nikolaou, Sergios Gatidis, Bin Yang. Comput. Medical Imaging Graph., 2018

709 citations recorded by the source corpus.

Image-to-image translation is considered a new frontier in the field of medical image analysis, with numerous potential applications. However, a large portion of recent approaches offers individualized solutions based on specialized task-specific architectures or require refinement through non-end-to-end training. In this paper, we propose a new framework, named MedGAN, for medical image-to-image translation which operates on the image level in an end-to-end manner. MedGAN builds upon recent advances in the field of generative adversarial networks (GANs) by merging the adversarial framework with a new combination of non-adversarial losses. We utilize a discriminator network as a trainable feature extractor which penalizes the discrepancy between the translated medical images and the desired modalities. Moreover, style-transfer losses are utilized to match the textures and fine-structures of the desired target images to the translated images. Additionally, we present a new generator architecture, titled CasNet, which enhances the sharpness of the translated medical outputs through progressive refinement via encoder-decoder pairs. Without any application-specific modifications, we apply MedGAN on three different tasks: PET-CT translation, correction of MR motion artefacts and PET image denoising. Perceptual analysis by radiologists and quantitative evaluations illustrate that the MedGAN outperforms other existing translation approaches.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/15d113b2cd4eb20ac8f26d1169bb7223045975a3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Synthetic Data for Deep Learning (2019, methodology): “Similar image-to-image translation techniques have been applied to generating images from 2D MR brain images to CT and back [282, 424, 639], PET to CT [27], cardiac CT to MR [96], virtual”
Self-derived organ attention for unpaired CT-MRI deep domain adaptation based MRI segmentation (2020, methodology): “Improved I2I translations have been obtained by adding style and perceptual losses for computing cyclical consistency (Armanious et al 2020) or with feature disentanglement methods (Yang et al 2019).”
Unified Focal loss: Generalising Dice and cross entropy-based losses to handle class imbalanced medical image segmentation (2021, background): “…al., 2020; L. Liu et al., 2020; Rundo et al., 2019a) — including the 3D U-Net (Cicek et al., 2016), Attention U-Net (Schlemper et al., 2019) and V-Net (Mil-letari et al., 2016) — as well as integrated into conditional Generative Adversarial Networks (Kessler et al., 2020; Armanious et al., 2020).”
A Survey on Generative Adversarial Networks: Variants, Applications, and Training (2020, methodology): “Furthermore, GANs have also been utilized in many other medical applications such as drug discovery GAN (Chem-GAN) [88] for the determination of internal chemical diversity of drugs, feedback GAN (FB-GAN) [89] for generating and designing deoxyribonucleic acid (DNA), medical image processing [90] fo”
Manipulating Medical Image Translation with Manifold Disentanglement (2020, background): “The most common application of conditional GAN is image domain translation [19].”
Generative Adversarial Network in Medical Imaging: A Review (2018, methodology): “(2017b) pix2pix* L1, 2 – M22 [é] [3D] Yi and Babyn (2018) pix2pix* L1, 2, 6 D1 M10, 11, 18, 19 [Ë] Yang et al. (2017b) pix2pix* L1, 2, 8 D2 M10, 11, 19 [Ë] [3D] Abdomen Kang et al. (2018) CycleGAN* L1, 3, 18 – M10, 19 [é] Coronary CT You et al. (2018) pix2pix* L1, 2, 9 D2 M1, 9, 10, 11, 19 [Ë] [3D] ”
Image-to-Image Translation: Methods and Applications (2021, background): “image denoising [192] \ [193], [194] \ [103] \ \ \ \ \ \”
A review of deep learning in medical imaging: Imaging traits, technology trends, case studies with progress highlights, and future promises (2020, methodology): “To help facilitate such work, the popular MedGAN [156] proposes a series of modifications and new loss functions to traditional GANs, aimed at preserving anatomically relevant information and fine details.”
Generation and evaluation of synthetic patient data (2020, background): “They have proven to be very adept at learning high-dimensional, continuous data such as images [26, 27].”