Toggle Main Menu Toggle Search

Open Access padlockePrints

Robust 3D U-Net Segmentation of Macular Holes

Lookup NU author(s): Dr Chris Willcocks, Maged Habib, Professor David SteelORCiD, Professor Boguslaw ObaraORCiD



This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).


© 2021 CEUR-WS. All rights reserved.Macular holes are a common eye condition which result in visual impairment. We look at the application of deep convolutional neural networks to the problem of macular hole segmentation. We use the 3D U-Net architecture as a basis and experiment with a number of design variants. Manually annotating and measuring macular holes is time consuming and error prone, taking dozens of minutes to annotate a single 3D scan. Previous automated approaches to macular hole segmentation take minutes to segment a single 3D scan. We found that, in less than one second, deep learning models generate significantly more accurate segmentations than previous automated approaches (Jaccard index boost of 0.08 - 0.09) and expert agreement (Jaccard index boost of 0.13 - 0.20). We also demonstrate that an approach of architectural simplification, by greatly simplifying the network capacity and depth, results in a model which is competitive with state-of-the-art models such as residual 3D U-Nets.

Publication metadata

Author(s): Frawley J, Willcocks CG, Habib M, Geenen C, Steel DH, Obara B

Editor(s): Arjun Pakrashi, Ellen Rushe, Mehran Hossein Zadeh Bazargani, Brian Mac Namee

Publication type: Conference Proceedings (inc. Abstract)

Publication status: Published

Conference Name: The 29th Irish Conference on Artificial Intelligence and Cognitive Science (AICS 2021)

Year of Conference: 2021

Pages: 36-47

Online publication date: 09/12/2021

Acceptance date: 02/04/2018

Date deposited: 29/03/2022

ISSN: 1613-0073

Publisher: CEUR-WS


Series Title: CEUR Workshop Proceedings