- -

Retinal Image Synthesis and Semi-supervised Learning for Glaucoma Assessment

RiuNet: Institutional repository of the Polithecnic University of Valencia

Share/Send to

Cited by

Statistics

Retinal Image Synthesis and Semi-supervised Learning for Glaucoma Assessment

Show simple item record

Files in this item

dc.contributor.author Díaz-Pinto, Andrés Yesid es_ES
dc.contributor.author Colomer, Adrián es_ES
dc.contributor.author Naranjo Ornedo, Valeriana es_ES
dc.contributor.author Morales, Sandra es_ES
dc.contributor.author Xu, Yanwu es_ES
dc.contributor.author Frangi, Alejandro F. es_ES
dc.date.accessioned 2019-07-26T20:01:04Z
dc.date.available 2019-07-26T20:01:04Z
dc.date.issued 2019 es_ES
dc.identifier.issn 0278-0062 es_ES
dc.identifier.uri http://hdl.handle.net/10251/124279
dc.description.abstract [EN] Recent works show that Generative Adversarial Networks (GANs) can be successfully applied to image synthesis and semi-supervised learning, where, given a small labelled database and a large unlabelled database, the goal is to train a powerful classifier. In this paper, we trained a retinal image synthesizer and a semi-supervised learning method for automatic glaucoma assessment using an adversarial model on a small glaucoma-labelled and large unlabelled database. Various studies have shown that glaucoma can be monitored by analyzing the optic disc and its surroundings, for that reason the images used in this work were automatically cropped around the optic disc. The novelty of this work is to propose a new retinal image synthesizer and a semi-supervised learning method for glaucoma assessment based on the Deep Convolutional Generative Adversarial Networks (DCGAN). In addition, and to the best of the authors' knowledge, this system is trained on an unprecedented number of publicly available images (86926 images). This system, hence, is not only able to generate images synthetically but to provide labels automatically. Synthetic images were qualitatively evaluated using t-SNE plots of features associated with the images and their anatomical consistency were estimated by measuring the proportion of pixels corresponding to the anatomical structures around the optic disc. The resulting image synthesizer is able to generate realistic (cropped) retinal images and, subsequently, the glaucoma classifier is able to classify them into glaucomatous and normal with high accuracy (AUC=0.9017). The obtained retinal image synthesizer and the glaucoma classifier could be used then to generate an unlimited number of cropped retinal images with glaucoma labels. es_ES
dc.language Inglés es_ES
dc.publisher Institute of Electrical and Electronics Engineers es_ES
dc.relation MINECO/BES-2014-067889 es_ES
dc.relation GV/GRISOLIA/2015/027
dc.relation.ispartof IEEE Transactions on Medical Imaging es_ES
dc.rights Reserva de todos los derechos es_ES
dc.subject Glaucoma Assessment es_ES
dc.subject Retinal Image Synthesis es_ES
dc.subject Fundus Images es_ES
dc.subject DCGAN es_ES
dc.subject Medical imaging es_ES
dc.subject.classification TEORIA DE LA SEÑAL Y COMUNICACIONES es_ES
dc.title Retinal Image Synthesis and Semi-supervised Learning for Glaucoma Assessment es_ES
dc.type Artículo es_ES
dc.identifier.doi 10.1109/TMI.2019.2903434 es_ES
dc.relation.projectID info:eu-repo/grantAgreement/EC/H2020/732613/EU es_ES
dc.rights.accessRights Abierto es_ES
dc.contributor.affiliation Universitat Politècnica de València. Departamento de Comunicaciones - Departament de Comunicacions es_ES
dc.description.bibliographicCitation Díaz-Pinto, AY.; Colomer, A.; Naranjo Ornedo, V.; Morales, S.; Xu, Y.; Frangi, AF. (2019). Retinal Image Synthesis and Semi-supervised Learning for Glaucoma Assessment. IEEE Transactions on Medical Imaging. 1-8. https://doi.org/10.1109/TMI.2019.2903434 es_ES
dc.description.accrualMethod S es_ES
dc.relation.publisherversion https://doi.org/10.1109/TMI.2019.2903434 es_ES
dc.description.upvformatpinicio 1 es_ES
dc.description.upvformatpfin 8 es_ES
dc.type.version info:eu repo/semantics/publishedVersion es_ES
dc.relation.pasarela 377539 es_ES
dc.contributor.funder MINISTERIO DE ECONOMIA Y EMPRESA es_ES
dc.contributor.funder Generalitat Valenciana (GV)


This item appears in the following Collection(s)

Show simple item record