Please use this identifier to cite or link to this item:
Title: Medical image editing in the latent space of Generative Adversarial Networks
Author: Fernández, Rubén
Rosado, Pilar
Vegas Lozano, Esteban
Reverter Comes, Ferran
Keywords: Intel·ligència artificial en medicina
Aprenentatge automàtic
Tècniques histològiques
Imatges mèdiques
Medical artificial intelligence
Machine learning
Histological techniques
Imaging systems in medicine
Issue Date: 28-Jul-2021
Publisher: Elsevier B.V.
Abstract: We consider a set of arithmetic operations in the latent space of Generative Adversarial Networks (GANs) to edit histopathological images. We analyze thousands of image patches from whole-slide images of breast cancer metastases in histological lymph node sections. Image files were downloaded from the pathology contests CAMELYON 16 and 17. We show that widely known architectures, such as: Deep Convolutional Generative Adversarial Networks (DCGAN) and Conditional Deep Convolutional Generative Adversarial Networks (cDCGAN), allow image editing using semantic concepts that represent underlying visual patterns in histopathological images, expanding GAN's well-known capabilities in medical image editing. We computed the Grad-cam heatmap of real positive images and of generated positive images, validating that the highlighted features both in the real and synthetic images match. We also show that GANs can be used to generate quality images, making GANs a valuable resource for augmenting small medical imaging datasets.
Note: Reproducció del document publicat a:
It is part of: Intelligence-Based Medicine, 2021, vol. 5
Related resource:
ISSN: 2666-5212
Appears in Collections:Articles publicats en revistes (Genètica, Microbiologia i Estadística)

Files in This Item:
File Description SizeFormat 
713572.pdf18.52 MBAdobe PDFView/Open

This item is licensed under a Creative Commons License Creative Commons