Study of Generative Adversarial Networks for Generating Synthetic Data and its Application on Optoacoustic Tomography

This work proposes the use of a Generative Adversarial Network (GAN) to perform data augmentation with the goal of improving image reconstruction in Optoacustic Tomography (OAT) applications. We employ the FastGAN model, a compact net capable of generating high resolution images from small datasets....

Full description

Bibliographic Details
Main Authors: Alejandro Scopa Lopina, Martín Germán González, Matías Vera
Format: Article
Language:English
Published: Universidad de Buenos Aires 2023-12-01
Series:Revista Elektrón
Subjects:
Online Access:http://elektron.fi.uba.ar/index.php/elektron/article/view/185
_version_ 1797389060386324480
author Alejandro Scopa Lopina
Martín Germán González
Matías Vera
author_facet Alejandro Scopa Lopina
Martín Germán González
Matías Vera
author_sort Alejandro Scopa Lopina
collection DOAJ
description This work proposes the use of a Generative Adversarial Network (GAN) to perform data augmentation with the goal of improving image reconstruction in Optoacustic Tomography (OAT) applications. We employ the FastGAN model, a compact net capable of generating high resolution images from small datasets. The quality of the generated data was assessed by two methods. First, the Fréchet distance (FID) was measured, observing a decreasing trend throughout the entire GAN training. Then, a U-Net neural network designed for a OAT system with and without augmented data was trained. In this case, the model trained with the extra data generated by the GAN achieved an appreciable improvement in the figures of merit associated with the reconstruction.
first_indexed 2024-03-08T22:50:20Z
format Article
id doaj.art-1d5f56d3be5042e58bb3165e98bbdd38
institution Directory Open Access Journal
issn 2525-0159
language English
last_indexed 2024-03-08T22:50:20Z
publishDate 2023-12-01
publisher Universidad de Buenos Aires
record_format Article
series Revista Elektrón
spelling doaj.art-1d5f56d3be5042e58bb3165e98bbdd382023-12-16T16:14:24ZengUniversidad de Buenos AiresRevista Elektrón2525-01592023-12-0172617010.37537/rev.elektron.7.2.185.2023109Study of Generative Adversarial Networks for Generating Synthetic Data and its Application on Optoacoustic TomographyAlejandro Scopa Lopina0Martín Germán González1Matías Vera2Universidad de Buenos AiresFIUBAFIUBAThis work proposes the use of a Generative Adversarial Network (GAN) to perform data augmentation with the goal of improving image reconstruction in Optoacustic Tomography (OAT) applications. We employ the FastGAN model, a compact net capable of generating high resolution images from small datasets. The quality of the generated data was assessed by two methods. First, the Fréchet distance (FID) was measured, observing a decreasing trend throughout the entire GAN training. Then, a U-Net neural network designed for a OAT system with and without augmented data was trained. In this case, the model trained with the extra data generated by the GAN achieved an appreciable improvement in the figures of merit associated with the reconstruction.http://elektron.fi.uba.ar/index.php/elektron/article/view/185tomografía optoacústicaaprendizaje profundoredes generativas de confrontacióndatos sintéticos
spellingShingle Alejandro Scopa Lopina
Martín Germán González
Matías Vera
Study of Generative Adversarial Networks for Generating Synthetic Data and its Application on Optoacoustic Tomography
Revista Elektrón
tomografía optoacústica
aprendizaje profundo
redes generativas de confrontación
datos sintéticos
title Study of Generative Adversarial Networks for Generating Synthetic Data and its Application on Optoacoustic Tomography
title_full Study of Generative Adversarial Networks for Generating Synthetic Data and its Application on Optoacoustic Tomography
title_fullStr Study of Generative Adversarial Networks for Generating Synthetic Data and its Application on Optoacoustic Tomography
title_full_unstemmed Study of Generative Adversarial Networks for Generating Synthetic Data and its Application on Optoacoustic Tomography
title_short Study of Generative Adversarial Networks for Generating Synthetic Data and its Application on Optoacoustic Tomography
title_sort study of generative adversarial networks for generating synthetic data and its application on optoacoustic tomography
topic tomografía optoacústica
aprendizaje profundo
redes generativas de confrontación
datos sintéticos
url http://elektron.fi.uba.ar/index.php/elektron/article/view/185
work_keys_str_mv AT alejandroscopalopina studyofgenerativeadversarialnetworksforgeneratingsyntheticdataanditsapplicationonoptoacoustictomography
AT martingermangonzalez studyofgenerativeadversarialnetworksforgeneratingsyntheticdataanditsapplicationonoptoacoustictomography
AT matiasvera studyofgenerativeadversarialnetworksforgeneratingsyntheticdataanditsapplicationonoptoacoustictomography