Full metadata record

DC Field Value Language
dc.contributor.author Jung, Euijin ko
dc.contributor.author Luna, Acevedo Miguel Andres ko
dc.contributor.author Park, Sang Hyun ko
dc.date.accessioned 2021-01-29T07:23:20Z -
dc.date.available 2021-01-29T07:23:20Z -
dc.date.created 2020-10-29 -
dc.date.issued 2020-10-05 -
dc.identifier.citation International Conference on Medical Image Computing and Computer Assisted Interventions, pp.79 - 90 -
dc.identifier.isbn 9783030593537 -
dc.identifier.issn 0302-9743 -
dc.identifier.uri http://hdl.handle.net/20.500.11750/12875 -
dc.description.abstract Predicting the evolution of Alzheimer’s disease (AD) is important for accurate diagnosis and the development of personalized treatments. However, learning a predictive model is challenging since it is difficult to obtain a large amount of data that includes changes over a long period of time. Conditional Generative Adversarial Networks (cGAN) may be an effective way to generate images that match specific conditions, but they are impractical to generate 3D images due to memory resource limitations. To address this issue, we propose a novel cGAN that is capable of synthesizing MR images at different stages of AD (i.e., normal, mild cognitive impairment, and AD). The proposed method consists of a 2D generator that synthesizes an image according to a condition with the help of 2D and 3D discriminators that evaluate how realistic the synthetic image is. We optimize both the 2D GAN loss and the 3D GAN loss to determine whether multiple consecutive 2D images generated in a mini-batch have real or fake appearance in 3D space. The proposed method can generate smooth and natural 3D images at different conditions by using a single network without large memory requirements. Experimental results show that the proposed method can generate better quality 3D MR images than 2D or 3D cGAN and can also boost the classification performance when the synthesized images are used to train a classification model. © 2020, Springer Nature Switzerland AG. -
dc.language English -
dc.publisher Springer Science and Business Media Deutschland GmbH -
dc.title Conditional Generative Adversarial Network for Predicting 3D Medical Images Affected by Alzheimer’s Diseases -
dc.type Conference -
dc.identifier.doi 10.1007/978-3-030-59354-4_8 -
dc.identifier.scopusid 2-s2.0-85092925551 -
dc.type.local Article(Overseas) -
dc.type.rims CONF -
dc.description.journalClass 1 -
dc.contributor.localauthor Park, Sang Hyun -
dc.identifier.citationStartPage 79 -
dc.identifier.citationEndPage 90 -
dc.identifier.citationTitle International Conference on Medical Image Computing and Computer Assisted Interventions -
dc.identifier.conferencecountry PE -
dc.identifier.conferencelocation Lima -

qrcode mendeley

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.