Generative adversarial networks for robust medical image analysis
Deep Learning models have been widely used for medical imaging tasks such as segmentation. However, these models tend to have low performances when applied to images that do not resemble the training dataset distribution. Thus, the robustness of medical segmentation models can be affected by externa...
- Autores:
-
Escobar Palomeque, María Camila
- Tipo de recurso:
- Fecha de publicación:
- 2020
- Institución:
- Universidad de los Andes
- Repositorio:
- Séneca: repositorio Uniandes
- Idioma:
- eng
- OAI Identifier:
- oai:repositorio.uniandes.edu.co:1992/50942
- Acceso en línea:
- http://hdl.handle.net/1992/50942
- Palabra clave:
- Sistemas de imágenes en medicina
Sistemas de representación tridimensional
Segmentación de imagen
Aprendizaje automático (Inteligencia artificial)
Redes neurales (Computadores)
Inteligencia artificial
Ingeniería
- Rights
- openAccess
- License
- http://creativecommons.org/licenses/by-nc-nd/4.0/
id |
UNIANDES2_6dd0407fd8c195feae2da602074f6c34 |
---|---|
oai_identifier_str |
oai:repositorio.uniandes.edu.co:1992/50942 |
network_acronym_str |
UNIANDES2 |
network_name_str |
Séneca: repositorio Uniandes |
repository_id_str |
|
spelling |
Al consultar y hacer uso de este recurso, está aceptando las condiciones de uso establecidas por los autores.http://creativecommons.org/licenses/by-nc-nd/4.0/info:eu-repo/semantics/openAccesshttp://purl.org/coar/access_right/c_abf2Arbeláez Escalante, Pablo Andrésvirtual::511-1Escobar Palomeque, María Camilab492d867-2284-45e5-a6a8-61a61bb38a65400Valderrama Manrique, Mario AndrésHernández Hoyos, Marcela2021-08-10T18:04:33Z2021-08-10T18:04:33Z2020http://hdl.handle.net/1992/5094222949.pdfinstname:Universidad de los Andesreponame:Repositorio Institucional Sénecarepourl:https://repositorio.uniandes.edu.co/Deep Learning models have been widely used for medical imaging tasks such as segmentation. However, these models tend to have low performances when applied to images that do not resemble the training dataset distribution. Thus, the robustness of medical segmentation models can be affected by external factors such as the quality of the input image, or by synthetic modifications such as adversarial attacks. In this work we present two novel approaches to increase robustness in medical segmentation by using Generative Adversarial Networks. First, we present UltraGAN, a method to improve the robustness to quality of ultrasound segmentation. Second, we present MedRobGAN, a method to generate adversarial examples that can later be used in improving the adversarial robustness for various 3D segmentation tasks.Los modelos de aprendizaje profundo se han utilizado ampliamente para tareas de imágenes médicas como la segmentación. Sin embargo, estos modelos tienden a tener un rendimiento bajo cuando se aplican a imágenes que no se parecen a la distribución del conjunto de datos de entrenamiento. Por tanto, la robustez de los modelos de segmentación médica puede verse afectada por factores externos como la calidad de la imagen de entrada o por modificaciones sintéticas como los ataques adversarios. En este trabajo presentamos dos enfoques novedosos para aumentar la robustez en la segmentación médica mediante el uso de Redes Adversarias Generativas. Primero, presentamos UltraGAN, un método para mejorar la robustez a la calidad de la segmentación por ultrasonido. En segundo lugar, presentamos MedRobGAN, un método para generar ejemplos adversarios que luego se pueden utilizar para mejorar la robustez adversaria para varias tareas de segmentación 3D.Magíster en Ingeniería BiomédicaMaestría12 hojasapplication/pdfengUniversidad de los AndesMaestría en Ingeniería BiomédicaFacultad de IngenieríaDepartamento de Ingeniería BiomédicaGenerative adversarial networks for robust medical image analysisTrabajo de grado - Maestríainfo:eu-repo/semantics/masterThesishttp://purl.org/coar/version/c_970fb48d4fbd8a85Texthttp://purl.org/redcol/resource_type/TMSistemas de imágenes en medicinaSistemas de representación tridimensionalSegmentación de imagenAprendizaje automático (Inteligencia artificial)Redes neurales (Computadores)Inteligencia artificialIngeniería201423470Publicationhttps://scholar.google.es/citations?user=k0nZO90AAAAJvirtual::511-10000-0001-5244-2407virtual::511-1https://scienti.minciencias.gov.co/cvlac/visualizador/generarCurriculoCv.do?cod_rh=0001579086virtual::511-1b4f52d42-ce2a-4e74-a22f-e52a6bfbd48evirtual::511-1b4f52d42-ce2a-4e74-a22f-e52a6bfbd48evirtual::511-1THUMBNAIL22949.pdf.jpg22949.pdf.jpgIM Thumbnailimage/jpeg9924https://repositorio.uniandes.edu.co/bitstreams/9556d2ff-cc44-408b-9ffa-3041ed0e3e28/downloadbe6ff0a4bfd65dbcb6554d57cc718defMD55ORIGINAL22949.pdfapplication/pdf4419768https://repositorio.uniandes.edu.co/bitstreams/24f80e5b-30b9-4e04-8259-e8790fe7f99b/download770af2f47e07ff8b12dbb86151198631MD51TEXT22949.pdf.txt22949.pdf.txtExtracted texttext/plain50692https://repositorio.uniandes.edu.co/bitstreams/286e0b9d-5236-4ce6-950b-9b30357bc655/download27127b349d9f7ac989f4e08f6713a560MD541992/50942oai:repositorio.uniandes.edu.co:1992/509422024-03-13 11:44:18.923http://creativecommons.org/licenses/by-nc-nd/4.0/open.accesshttps://repositorio.uniandes.edu.coRepositorio institucional Sénecaadminrepositorio@uniandes.edu.co |
dc.title.spa.fl_str_mv |
Generative adversarial networks for robust medical image analysis |
title |
Generative adversarial networks for robust medical image analysis |
spellingShingle |
Generative adversarial networks for robust medical image analysis Sistemas de imágenes en medicina Sistemas de representación tridimensional Segmentación de imagen Aprendizaje automático (Inteligencia artificial) Redes neurales (Computadores) Inteligencia artificial Ingeniería |
title_short |
Generative adversarial networks for robust medical image analysis |
title_full |
Generative adversarial networks for robust medical image analysis |
title_fullStr |
Generative adversarial networks for robust medical image analysis |
title_full_unstemmed |
Generative adversarial networks for robust medical image analysis |
title_sort |
Generative adversarial networks for robust medical image analysis |
dc.creator.fl_str_mv |
Escobar Palomeque, María Camila |
dc.contributor.advisor.none.fl_str_mv |
Arbeláez Escalante, Pablo Andrés |
dc.contributor.author.none.fl_str_mv |
Escobar Palomeque, María Camila |
dc.contributor.jury.none.fl_str_mv |
Valderrama Manrique, Mario Andrés Hernández Hoyos, Marcela |
dc.subject.armarc.es_CO.fl_str_mv |
Sistemas de imágenes en medicina Sistemas de representación tridimensional Segmentación de imagen Aprendizaje automático (Inteligencia artificial) Redes neurales (Computadores) Inteligencia artificial |
topic |
Sistemas de imágenes en medicina Sistemas de representación tridimensional Segmentación de imagen Aprendizaje automático (Inteligencia artificial) Redes neurales (Computadores) Inteligencia artificial Ingeniería |
dc.subject.themes.none.fl_str_mv |
Ingeniería |
description |
Deep Learning models have been widely used for medical imaging tasks such as segmentation. However, these models tend to have low performances when applied to images that do not resemble the training dataset distribution. Thus, the robustness of medical segmentation models can be affected by external factors such as the quality of the input image, or by synthetic modifications such as adversarial attacks. In this work we present two novel approaches to increase robustness in medical segmentation by using Generative Adversarial Networks. First, we present UltraGAN, a method to improve the robustness to quality of ultrasound segmentation. Second, we present MedRobGAN, a method to generate adversarial examples that can later be used in improving the adversarial robustness for various 3D segmentation tasks. |
publishDate |
2020 |
dc.date.issued.none.fl_str_mv |
2020 |
dc.date.accessioned.none.fl_str_mv |
2021-08-10T18:04:33Z |
dc.date.available.none.fl_str_mv |
2021-08-10T18:04:33Z |
dc.type.spa.fl_str_mv |
Trabajo de grado - Maestría |
dc.type.coarversion.fl_str_mv |
http://purl.org/coar/version/c_970fb48d4fbd8a85 |
dc.type.driver.spa.fl_str_mv |
info:eu-repo/semantics/masterThesis |
dc.type.content.spa.fl_str_mv |
Text |
dc.type.redcol.spa.fl_str_mv |
http://purl.org/redcol/resource_type/TM |
dc.identifier.uri.none.fl_str_mv |
http://hdl.handle.net/1992/50942 |
dc.identifier.pdf.none.fl_str_mv |
22949.pdf |
dc.identifier.instname.spa.fl_str_mv |
instname:Universidad de los Andes |
dc.identifier.reponame.spa.fl_str_mv |
reponame:Repositorio Institucional Séneca |
dc.identifier.repourl.spa.fl_str_mv |
repourl:https://repositorio.uniandes.edu.co/ |
url |
http://hdl.handle.net/1992/50942 |
identifier_str_mv |
22949.pdf instname:Universidad de los Andes reponame:Repositorio Institucional Séneca repourl:https://repositorio.uniandes.edu.co/ |
dc.language.iso.none.fl_str_mv |
eng |
language |
eng |
dc.rights.uri.*.fl_str_mv |
http://creativecommons.org/licenses/by-nc-nd/4.0/ |
dc.rights.accessrights.spa.fl_str_mv |
info:eu-repo/semantics/openAccess |
dc.rights.coar.spa.fl_str_mv |
http://purl.org/coar/access_right/c_abf2 |
rights_invalid_str_mv |
http://creativecommons.org/licenses/by-nc-nd/4.0/ http://purl.org/coar/access_right/c_abf2 |
eu_rights_str_mv |
openAccess |
dc.format.extent.none.fl_str_mv |
12 hojas |
dc.format.mimetype.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Universidad de los Andes |
dc.publisher.program.none.fl_str_mv |
Maestría en Ingeniería Biomédica |
dc.publisher.faculty.none.fl_str_mv |
Facultad de Ingeniería |
dc.publisher.department.none.fl_str_mv |
Departamento de Ingeniería Biomédica |
publisher.none.fl_str_mv |
Universidad de los Andes |
institution |
Universidad de los Andes |
bitstream.url.fl_str_mv |
https://repositorio.uniandes.edu.co/bitstreams/9556d2ff-cc44-408b-9ffa-3041ed0e3e28/download https://repositorio.uniandes.edu.co/bitstreams/24f80e5b-30b9-4e04-8259-e8790fe7f99b/download https://repositorio.uniandes.edu.co/bitstreams/286e0b9d-5236-4ce6-950b-9b30357bc655/download |
bitstream.checksum.fl_str_mv |
be6ff0a4bfd65dbcb6554d57cc718def 770af2f47e07ff8b12dbb86151198631 27127b349d9f7ac989f4e08f6713a560 |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 MD5 |
repository.name.fl_str_mv |
Repositorio institucional Séneca |
repository.mail.fl_str_mv |
adminrepositorio@uniandes.edu.co |
_version_ |
1812133799344996352 |