GANs for Medical Image Analysis

By Salome Kazeminia, Christoph Baur, Arjan Kuijper, B. van Ginneken, Nassir Navab, Shadi Albarqouni, Anirban Mukhopadhyay. 2018

505 citations recorded by the source corpus.

Generative adversarial networks (GANs) and their extensions have carved open many exciting ways to tackle well known and challenging medical image analysis problems such as medical image de-noising, reconstruction, segmentation, data simulation, detection or classification. Furthermore, their ability to synthesize images at unprecedented levels of realism also gives hope that the chronic scarcity of labeled data in the medical field can be resolved with the help of these generative models. In this review paper, a broad overview of recent literature on GANs for medical applications is given, the shortcomings and opportunities of the proposed methods are thoroughly discussed, and potential future work is elaborated. We review the most relevant papers published until the submission date. For quick access, essential details such as the underlying method, datasets, and performance are tabulated. An interactive visualization that categorizes all papers to keep the review alive is available at http://livingreview.in.tum.de/GANs_for_Medical_Applications/.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

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

In Computer Vision and Pattern Recognition

Citations in the corpus

Style transfer-based image synthesis as an efficient regularization technique in deep learning (2019, background): “Recently, GANs started gaining attention in the field of medical image processing to solve tasks such as de-noising [29], segmentation [30], detection [31] and classification [32].”
How Artificial Intelligence Is Shaping Medical Imaging Technology: A Survey of Innovations and Applications (2023, methodology): “Generative adversarial networks (GANs) have been a promising tool in the field of medical image analysis [57], particularly in image-to-image translation.”
Recent advances and clinical applications of deep learning in medical image analysis (2021, result): “In contrast to the existing review papers (Kazeminia et al., 2020b; Yi et al., 2019), we reviewed the research progress of GANs from two aspects: 1) different loss (objective) functions, and 2) conditional and unconditional settings.”
Diffusion models in medical imaging: A comprehensive survey (2022, background): “Noise reduces the image quality and is especially significant when the imaged objects are small and have relatively low contrast [115].”
High-resolution Controllable Prostatic Histology Synthesis using StyleGAN (2021, methodology): “…Gleason patterns indicative of patient prognosis from histology (Young et al., 2019; Li et al., 2019; Araújo et al., 2017) and generative adversarial networks (GANs) that are able to reconstruct novel medical images from a latent code or label (Kazeminia et al., 2018; Karras et al., 2020).”
Deep learning in medical image registration: a survey (2019, background): “review articles that survey the application of deep learning to medical image analysis [59,63], reinforcement learning [51], and the application of GANs to medical image analysis [52] are recommended to the interested readers.”