Synthetic Medical Images from Dual Generative Adversarial Networks

By John Guibas, Tejpal S. Virdi, Peter S. Li. 2017

186 citations recorded by the source corpus.

Currently there is strong interest in data-driven approaches to medical image classification. However, medical imaging data is scarce, expensive, and fraught with legal concerns regarding patient privacy. Typical consent forms only allow for patient data to be used in medical journals or education, meaning the majority of medical data is inaccessible for general public research. We propose a novel, two-stage pipeline for generating synthetic medical images from a pair of generative adversarial networks, tested in practice on retinal fundi images. We develop a hierarchical generation process to divide the complex image generation task into two parts: geometry and photorealism. We hope researchers will use our pipeline to bring private medical data into the public domain, sparking growth in imaging tasks that have previously relied on the hand-tuning of models. We have begun this initiative through the development of SynthMed, an online repository for synthetic medical images.

Subfields: Artificial Intelligence, Biophysics, Computer Vision and Pattern Recognition, Generative Models, Medical Imaging

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 Biophysics

In Computer Vision and Pattern Recognition

In Generative Models

In Medical Imaging

Citations in the corpus

High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs (2017, neutral): “, medical imaging [17] and biology [8]).”
An overview of deep learning in medical imaging focusing on MRI (2018, methodology): “[214] used a two-stage pipeline for generating synthetic medical images from a pair of GANs, addressing retinal fundus images, and provided an online repository (SynthMed) for synthetic medical images.”
Generative Adversarial Network in Medical Imaging: A Review (2018, methodology): “With the training stability mproved, some works have also incorporated residual connections nto both the generator and discriminator and experimented with uch deeper networks ( Gulrajani et al., 2017; Miyato et al., 2018 ). he work in Miyato and Koyama (2018) proposed a projection ased way to incorp”
Deep Learning Applications in Medical Image Analysis (2018, methodology): “This was done by Guibas and Virdi, who used a 2 stage GAN to segment and then generate retinal fundus images successfully [89].”
GANs for Medical Image Analysis (2018, methodology): “e synthetic vessel tree images in turn can be fed into the image-to-image translation model, leading to an end-to-end framework for realistic, high resolution retinal image synthesis. Very similarly, Guibas et al. (2017) propose a two-stage approach, consisting of a GAN trained to synthesize vessel ”
Ensembling with Deep Generative Views (2021, background): “Specifically in the medical community, synthesizing additional images to augment limited-size datasets has been beneficial for classification [40, 60, 17, 22, 48] or segmentation tasks [32, 11, 20].”
Unsupervised Reverse Domain Adaptation for Synthetic Medical Images via Adversarial Training (2017, background): “In the medical imaging domain, there has been recent success in generating realistic synthetic data for the relatively constrained problem of 2D retinal imaging using standard GANs [16], [17].”
Manipulating Medical Image Translation with Manifold Disentanglement (2020, background): “For example, segmentation map to medical image [20], [21], MR reconstruction [22], [23], image denoising [24], [25] and cross-modality translation [26], [27], [28].”
Generative Adversarial Networks in Time Series: A Systematic Literature Review (2022, background): “Since their inception in 2014, GANs have shown great success in generating high-quality synthetic images indistinguishable from real images [41, 65, 87].”
Synthesis of Realistic ECG using Generative Adversarial Networks (2019, background): “Since their inception in 2014, GANs have shown great success in generating high-quality synthetic images which are indistinguishable from the real images [21, 5, 6].”
Embracing Imperfect Datasets: A Review of Deep Learning Solutions for Medical Image Segmentation (2019, methodology): “Via conditional GANs: Guibas et al. (2017) propose a framework consisting of a GAN and a conditional GAN to generate pairs of synthetic fundus images and the corresponding vessel masks.”