Banca de DEFESA: EDUARDO FELIPE DE SOUZA



Uma banca de DEFESA DE MESTRADO foi cadastrada pelo programa.

DISCENTE: EDUARDO FELIPE DE SOUZA
DATA: 23/11/2020
HORA: 15:00
LOCAL: Google Meet
TÍTULO:

MMI-GAN: Multi Medical Imaging Translation using GenerativeAdversarial Network


RESUMO:

Medical image translation is considered a new frontier in the field of medical image analysis, with great potential for application. However, the existing approaches have limited scalability and robustness in handling more than two domains of images, since different models must be created independently for each pair of domains. In addition, there are problems with the quality of the translated images. To solve these limitations, we developed MMI-GAN, a new approach for translation between multiple image domains, capable of translating intermodal (CT and RM) and intramodal (PD, T1 and T2) images. We demonstrate the efficiency of the conditional in learning to map between several domains using only a single generator and a discriminator, training data from images of all domains. We propose a GAN architecture that can be easily extended to other translation tasks for the benefit of the medical imaging community. The MMI-GAN is based on recent advances in the area of GANs (Generative Adversarial Network), using an adversary structure with a new combination of non-adversarial losses, which allows the simultaneous training of several data sets with different domains in the same network, as well as the innovative capacity to translate with flexibility between and intramodalities.


PALAVRAS-CHAVE:

Generative Adversarial Networks, image translation, multi-domain, magnetic resonance, computed tomography


PÁGINAS: 100
GRANDE ÁREA: Ciências Exatas e da Terra
ÁREA: Ciência da Computação
SUBÁREA: Metodologia e Técnicas da Computação
ESPECIALIDADE: Processamento Gráfico (Graphics)

MEMBROS DA BANCA:
Presidente - 1544992 - MARCELO COSTA OLIVEIRA
Interno(a) - 2087442 - TIAGO FIGUEIREDO VIEIRA
Externo(a) à Instituição - PAULO MAZZONCINI DE AZEVEDO MARQUES - USP
Notícia cadastrada em: 23/11/2020 13:14
SIGAA | NTI - Núcleo de Tecnologia da Informação - (82) 3214-1015 | Copyright © 2006-2024 - UFAL - sig-app-4.srv4inst1 30/04/2024 12:30