Self-supervised Deep-Learning Segmentation of Corneal Endothelium Specular Microscopy Images

Sergio Sanchez, Kevin Mendoza, Fernando Quintero, Angelica M. Prada, Alejandro Tello, Virgilio Galvis, Lenny A. Romero, Andres G. Marrugo

Producción científica: Libro / Capitulo del libro / InformeLibros de Investigaciónrevisión exhaustiva

Resumen

Computerized medical evaluation of the corneal endothelium is challenging because it requires costly equipment and specialized personnel, not to mention that conventional techniques require manual annotations that are difficult to acquire. This study aims to obtain reliable segmentations without requiring large data sets labeled by expert personnel. To address this problem, we use the Barlow Twins approach to pre-train the encoder of a UNet model in an unsupervised manner. Then, with few labeled data, we train the segmentation. Encouraging results show that it is possible to address the challenge of limited data availability using self-supervised learning. This model achieved a precision of 86%, obtaining a satisfactory performance. Using many images to learn good representations and a few labeled images to learn the semantic segmentation task is feasible.

Idioma originalInglés
Título de la publicación alojadaApplications of Computational Intelligence - 6th IEEE Colombian Conference, ColCACI 2023, Revised Selected Papers
EditoresAlvaro David Orjuela-Cañón, Jesus A Lopez, Julián David Arias-Londoño
EditorialSpringer Science and Business Media Deutschland GmbH
Páginas30-42
Número de páginas13
ISBN (versión impresa)9783031484148
DOI
EstadoPublicada - 2024
Evento6th IEEE Colombian Conference on Applications of Computational Intelligence, ColCACI 2023 - Bogota, Colombia
Duración: 26 jul. 202328 jul. 2023

Serie de la publicación

NombreCommunications in Computer and Information Science
Volumen1865 CCIS
ISSN (versión impresa)1865-0929
ISSN (versión digital)1865-0937

Conferencia

Conferencia6th IEEE Colombian Conference on Applications of Computational Intelligence, ColCACI 2023
País/TerritorioColombia
CiudadBogota
Período26/07/2328/07/23

Huella

Profundice en los temas de investigación de 'Self-supervised Deep-Learning Segmentation of Corneal Endothelium Specular Microscopy Images'. En conjunto forman una huella única.

Citar esto