PathologyGAN: Learning deep representations of cancer tissue

Claudio Quiros, A., Murray-Smith, R. and Yuan, K. (2021) PathologyGAN: Learning deep representations of cancer tissue. Journal of Machine Learning for Biomedical Imaging, 2021(4), pp. 1-48.

[img] Text
241563.pdf - Published Version
Available under License Creative Commons Attribution.

15MB

Publisher's URL: https://www.melba-journal.org/article/21657

Abstract

Histopathological images of tumours contain abundant information about how tumours grow and how they interact with their micro-environment. Better understanding of tissue phenotypes in these images could reveal novel determinants of pathological processes underlying cancer, and in turn improve diagnosis and treatment options. Advances of Deep learning makes it ideal to achieve those goals, however, its application is limited by the cost of high quality labels from patients data. Unsupervised learning, in particular, deep generative models with representation learning properties provides an alternative path to further understand cancer tissue phenotypes, capturing tissue morphologies. In this paper, we develop a framework which allows Generative Adversarial Networks (GANs) to capture key tissue features and uses these characteristics to give structure to its latent space. To this end, we trained our model on two different datasets, an H and E colorectal cancer tissue from the National Center for Tumor diseases (NCT, Germany) and an H and E breast cancer tissue from the Netherlands Cancer Institute (NKI, Netherlands) and Vancouver General Hospital (VGH, Canada). Composed of 86 slide images and 576 tissue micro-arrays (TMAs) respectively. We show that our model generates high quality images, with a Frechet Inception Distance (FID) of 16.65 (breast cancer) and 32.05 (colorectal cancer). We further assess the quality of the images with cancer tissue characteristics (e.g. count of cancer, lymphocytes, or stromal cells), using quantitative information to calculate the FID and showing consistent performance of 9.86. Additionally, the latent space of our model shows an interpretable structure and allows semantic vector operations that translate into tissue feature transformations. Furthermore, ratings from two expert pathologists found no significant difference between our generated tissue images from real ones. The code, generated images, and pretrained model are available at https://github.com/AdalbertoCq/Pathology-GAN

Item Type:Articles
Additional Information:Special Issue: Medical Imaging with Deep Learning (MIDL) 2020.
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Yuan, Dr Ke and Murray-Smith, Professor Roderick and Claudio Quiros, Mr Adalberto
Authors: Claudio Quiros, A., Murray-Smith, R., and Yuan, K.
College/School:College of Science and Engineering > School of Computing Science
Journal Name:Journal of Machine Learning for Biomedical Imaging
Publisher:MELBA
ISSN:2766-905X
ISSN (Online):2766-905X
Copyright Holders:Copyright © 2021 Adalberto Claudio Quiros, Roderick Murray-Smith, and Ke Yuan
First Published:First published in Journal of Machine Learning for Biomedical Imaging 2021(4):1-48
Publisher Policy:Reproduced under a Creative Commons Licence
Related URLs:

University Staff: Request a correction | Enlighten Editors: Update this record

Project CodeAward NoProject NamePrincipal InvestigatorFunder's NameFunder RefLead Dept
300982Exploiting Closed-Loop Aspects in Computationally and Data Intensive AnalyticsRoderick Murray-SmithEngineering and Physical Sciences Research Council (EPSRC)EP/R018634/1Computing Science
305567QuantIC - The UK Quantum Technoogy Hub in Quantum Enhanced ImagingMiles PadgettEngineering and Physical Sciences Research Council (EPSRC)EP/T00097X/1P&S - Physics & Astronomy
300982Exploiting Closed-Loop Aspects in Computationally and Data Intensive AnalyticsRoderick Murray-SmithEngineering and Physical Sciences Research Council (EPSRC)EP/R018634/1Computing Science