Super-resolution of magnetic resonance fetal images using sequential generative adversarial networks

Mostra el registre complet Registre parcial de l'ítem

• dc.contributor.author Obelleiro Liz, Manuel
• dc.date.accessioned 2022-10-26T14:24:54Z
• dc.date.available 2022-10-26T14:24:54Z
• dc.date.issued 2022
• dc.description Tutor: Benjamin Lalande Chatain
• dc.description Treball de fi de grau en Biomèdica
• dc.description.abstract Magnetic Resonance imaging is a really important imaging technique distinguished by the lack of ionizing radiation, great soft-tissue contrast and remarkable resolution. The main drawback of performing a Magnetic Resonance Imaging (MRI) scan is time consumption. As a result, many MRI scans are performed in a shorter time sacrificing overall image quality. Thus, there is a need for algorithms able to convert low-resolution images, obtained from faster scans, to super-resolution images comparable to full-time scans. Generative Adversarial Networks (GANs) have the ability to generate new content, mainly images and data distributions. This kind of network can be divided into 2 units that act in competition, a generator and a discriminator. The generator receives an input and must return, an output which follows the target distribution. The discriminator acts as the judge of the process, determining which image is real and which was created by the generator. In this bachelor thesis, a two-stage training process is developed using the ESRGAN framework. First, a Peak Signal to Noise Ratio (PSNR) oriented model is developed without adversarial training, performing similar to standard deep networks. In the second stage, the discriminator is introduced and trained together with the PSNR-oriented generator. The second stage allows the model to learn realistic texture representations, measured with the perceptual index (P.I.) metric. The obtained results demonstrate the potential application of these networks to medical imaging. The reconstructed MRI images were objectively more similar to the original High Resolution (HR) ground truth when compared to other conventional image reconstruction methods. The PSNR trained model yielded better PSNR and structural similarity (SSIM) performance but with a lower perceptual index. The GAN model showed greater P.I. by being able to replicate image textures and creating overall more visually appealing images.ca
• dc.format.mimetype application/pdf*
• dc.identifier.uri http://hdl.handle.net/10230/54606
• dc.language.iso engca
• dc.rights ©Tots els drets reservatsca
• dc.rights.accessRights info:eu-repo/semantics/openAccessca
• dc.subject.keyword MRI
• dc.subject.keyword Fetal images
• dc.subject.keyword Image processing
• dc.subject.keyword Computer-aided diagnosis (CAD)
• dc.subject.keyword Single image super-resolution
• dc.subject.keyword Artificial intelligence (AI)
• dc.subject.keyword Deep learning (DL)
• dc.subject.keyword Deep convolutional neural networks
• dc.subject.keyword Generative adversarial networks (GANs)
• dc.title Super-resolution of magnetic resonance fetal images using sequential generative adversarial networksca
• dc.type info:eu-repo/semantics/bachelorThesisca