Central pit in the middle of the macula of the retina. Composed of closely packed cones, it is responsible for approximately half of the visual information produced by the whole retina.
Published in Chapter:
Fully Automatic Epiretinal Membrane Segmentation in OCT Scans Using Convolutional Networks
Mateo Gende (CITIC, INIBIC, Universidade da Coruña, Spain), Joaquim de Moura (CITIC, INIBIC, Universidade da Coruña, Spain), Jorge Novo (CITIC, INIBIC, Universidade da Coruña, Spain), and Marcos Ortega (CITIC, INIBIC, Universidade da Coruña, Spain)
Copyright: © 2022
|Pages: 34
DOI: 10.4018/978-1-6684-2304-2.ch004
Abstract
The epiretinal membrane (ERM) is an ocular pathology that can cause visual distortions. To prevent a loss of vision, symptomatic ERM needs to be removed before it can cause irreversible damage. In order to do this, the ERM needs to be located early, so that it can be peeled from the retina. This chapter explores an automatic methodology for ERM segmentation, as well as its intuitive visualization in the form of colour maps. To do this, visual features that are compatible with ERM presence are extracted from ophthalmologic images by using computer vision algorithms and deep learning models. This methodology achieved satisfactory results, reaching a dice coefficient of 0.826 and a Jaccard index of 0.714, contributing to highlight the applicability of deep learning models for the detection of pathological signs in medical images.