Impact of Image Complexity on the Segmentation of Ophthalmic Images Using Deep Learning Methods

Ignacio A. Viedma; David Alonso-Caneiro; Scott A. Read; Jason Charng; Fred K. Chen; David A. Mackey; Michael J. Collins

doi:10.53941/jbiop.2025.100002

Back

Impact of Image Complexity on the Segmentation of Ophthalmic Images Using Deep Learning Methods

Journal article

Open access

Peer reviewed

Impact of Image Complexity on the Segmentation of Ophthalmic Images Using Deep Learning Methods

Ignacio A. Viedma, David Alonso-Caneiro, Scott A. Read, Jason Charng, Fred K. Chen, David A. Mackey and Michael J. Collins

Journal of Bio-optics, Vol.1(1), pp.1-15

2025

DOI: https://doi.org/10.53941/jbiop.2025.100002

Files and links (1)

pdf

25040005181.73 MBDownload View

Published VersionCC BY V4.0, Open Access

Abstract

opital coherence tomography

scanning laser ophthalmoscopy

semantic segmentation

deep learning

image complexity

Advances in medical imaging segmentation using deep learning (DL) have facilitated the development of a wide range of models based on different architectures. For example, the U-Net has become one of the most widely used architectures in the field. Due to its popularity, various modifications to the original U-Net architecture have been proposed, with the aim to improve the segmentation performance. Most studies utilizing ophthalmic images, such as optical coherence tomography (OCT) and scanning laser ophthalmoscopy (SLO), have employed U-Net methods and their variations for segmentation. Given the multitude of U-Net variations, selecting the optimum model for ophthalmic image segmentation may be challenging, and other factors that may impact the model’s segmentation performance, such as image complexity and its impact on model selection, have been largely unexplored. Thus, in this study, the segmentation performance of three models, including a baseline U-Net, a popular U-Net variation (U-Net++), and a segmentation architecture with a different approach (DeepLabV3), are compared, analyzing how these different methods may vary for OCT and SLO datasets with various levels of segmentation task complexity. To analyze the effect of image complexity on segmentation performance, several metrics are extracted, including the image entropy of the datasets at the pixel level, the texture level, and the global features level. The results demonstrate a relationship between the complexity of the images in a dataset and the performance of the segmentation model used for the specific task. Data complexity may serve as a metric to inform DL model selection or aid in the early design process.

Details

Title: Impact of Image Complexity on the Segmentation of Ophthalmic Images Using Deep Learning Methods
Authors: Ignacio A. Viedma - Queensland University of Technology
David Alonso-Caneiro (Corresponding Author) - University of the Sunshine Coast, Queensland, School of Science, Technology and Engineering
Scott A. Read - Queensland University of Technology
Jason Charng - The University of Western Australia
Fred K. Chen - The University of Western Australia
David A. Mackey - The University of Western Australia
Michael J. Collins - Queensland University of Technology
Publication details: Journal of Bio-optics, Vol.1(1), pp.1-15
Publisher: Scilight
Date published: 2025
DOI: 10.53941/jbiop.2025.100002
ISSN: 3083-4651
Copyright note: Copyright (c) 2025 by the authors. This work is licensed under a Creative Commons Attribution 4.0 International License.
Data Availability: The dataset consisting of medical images is not publicly available at the time of writing due to institutional restrictions.
Grants: Predicting visual function from structural data in health and ocular disease, 1186915, National Health and Medical Research Council (Australia, Canberra) - NHMRC
Developing Personalised Treatment for Retinal Degeneration, 1142962, National Health and Medical Research Council (Australia, Canberra) - NHMRC
From discovery to therapy in genetic eye diseases, 1116360, National Health and Medical Research Council (Australia, Canberra) - NHMRC
Organisation Unit: School of Science, Technology and Engineering
Language: English
Record Identifier: 991166403202621
Output Type: Journal article

Metrics

1 File views/ downloads

5 Record Views