UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Adversarial training with cycle consistency for unsupervised super-resolution in endomicroscopy

Ravi, D; Szczotka, AB; Pereira, SP; Vercauteren, T; (2019) Adversarial training with cycle consistency for unsupervised super-resolution in endomicroscopy. Medical Image Analysis , 53 pp. 123-131. 10.1016/j.media.2019.01.011. Green open access

[thumbnail of Ravi_OA_1-s2.0-S1361841518305966-main.pdf]
Preview
Text
Ravi_OA_1-s2.0-S1361841518305966-main.pdf - Published version

Download (2MB) | Preview

Abstract

In recent years, endomicroscopy has become increasingly used for diagnostic purposes and interventional guidance. It can provide intraoperative aids for real-time tissue characterization and can help to perform visual investigations aimed for example to discover epithelial cancers. Due to physical constraints on the acquisition process, endomicroscopy images, still today have a low number of informative pixels which hampers their quality. Post-processing techniques, such as Super-Resolution (SR), are a potential solution to increase the quality of these images. SR techniques are often supervised, requiring aligned pairs of low-resolution (LR) and high-resolution (HR) images patches to train a model. However, in our domain, the lack of HR images hinders the collection of such pairs and makes supervised training unsuitable. For this reason, we propose an unsupervised SR framework based on an adversarial deep neural network with a physically-inspired cycle consistency, designed to impose some acquisition properties on the super-resolved images. Our framework can exploit HR images, regardless of the domain where they are coming from, to transfer the quality of the HR images to the initial LR images. This property can be particularly useful in all situations where pairs of LR/HR are not available during the training. Our quantitative analysis, validated using a database of 238 endomicroscopy video sequences from 143 patients, shows the ability of the pipeline to produce convincing super-resolved images. A Mean Opinion Score (MOS) study also confirms this quantitative image quality assessment.

Type: Article
Title: Adversarial training with cycle consistency for unsupervised super-resolution in endomicroscopy
Open access status: An open access version is available from UCL Discovery
DOI: 10.1016/j.media.2019.01.011
Publisher version: https://doi.org/10.1016/j.media.2019.01.011
Language: English
Additional information: © 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license. (http://creativecommons.org/licenses/by/4.0/.
Keywords: Deep learning, Probe-based confocal laser endomicroscopy, Unsupervised Super-resolution, Cycle consistency, Adversial training
UCL classification: UCL
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Medical Sciences
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Medical Sciences > Div of Medicine
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Medical Sciences > Div of Medicine > Inst for Liver and Digestive Hlth
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Med Phys and Biomedical Eng
URI: https://discovery.ucl.ac.uk/id/eprint/10066536
Downloads since deposit
56Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item