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...

Full description

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/
Description
Summary: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.