Effect of patch size and network architecture on a convolutional neural network approach for automatic segmentation of OCT retinal layers

Jared Hamwood; David Alonso-Caneiro; Scott A. Read; Stephen J. Vincent; Michael J. Collins

doi:10.1364/BOE.9.003049

Back

Effect of patch size and network architecture on a convolutional neural network approach for automatic segmentation of OCT retinal layers

Journal article

Open access

Peer reviewed

Effect of patch size and network architecture on a convolutional neural network approach for automatic segmentation of OCT retinal layers

Jared Hamwood, David Alonso-Caneiro, Scott A. Read, Stephen J. Vincent and Michael J. Collins

Biomedical Optics Express, Vol.9(7), pp.3049-3066

2018

DOI: https://doi.org/10.1364/BOE.9.003049

PMCID: PMC6033561

PMID: 29984082

Files and links (1)

pdf

boe-9-7-30495.15 MBDownload View

Published VersionCC BY V4.0, Open Access

Abstract

Deep learning strategies, particularly convolutional neural networks (CNNs), are especially suited to finding patterns in images and using those patterns for image classification. The method is normally applied to an image patch and assigns a class weight to the patch; this method has recently been used to detect the probability of retinal boundary locations in OCT images, which is subsequently used to segment the OCT image using a graph-search approach. This paper examines the effects of a number of modifications to the CNN architecture with the aim of optimizing retinal layer segmentation, specifically the effect of patch size as well as the network architecture design on CNN performance and subsequent layer segmentation. The results demonstrate that increasing patch size can improve the performance of the classification and provides a more reliable segmentation in the analysis of retinal layer characteristics in OCT imaging. Similarly, this work shows that changing aspects of the CNN network design can also significantly improve the segmentation results. This work also demonstrates that the performance of the method can change depending on the number of classes (i.e. boundaries) used to train the CNN, with fewer classes showing an inferior performance due to the presence of similar image features between classes that can trigger false positives. Changes in the network (patch size and or architecture) can be applied to provide a superior segmentation performance, which is robust to the class effect. The findings from this work may inform future CNN development in OCT retinal image analysis.

Details

Title: Effect of patch size and network architecture on a convolutional neural network approach for automatic segmentation of OCT retinal layers
Authors: Jared Hamwood (Author) - Queensland University of Technology
David Alonso-Caneiro (Corresponding Author) - Queensland University of Technology
Scott A. Read (Author) - Queensland University of Technology
Stephen J. Vincent (Author) - Queensland University of Technology
Michael J. Collins (Author) - Queensland University of Technology
Publication details: Biomedical Optics Express, Vol.9(7), pp.3049-3066
Publisher: Optica
Date published: 2018
DOI: 10.1364/BOE.9.003049
ISSN: 2156-7085
PMID: 29984082; PMC6033561
Copyright note: © 2018 Optical Society of America. Users may use, reuse, and build upon the article, or use the article for text or data mining, so long as such uses are for non-commercial purposes and appropriate attribution is maintained. All other rights are reserved.
Organisation Unit: School of Science, Technology and Engineering
Language: English
Record Identifier: 99972395402621
Output Type: Journal article

Metrics

24 Record Views

94 Times Cited - Web of Science

InCites Highlights

These are selected metrics from InCites Benchmarking & Analytics tool, related to this output

Web Of Science research areas: Biochemical Research Methods; Optics; Radiology, Nuclear Medicine & Medical Imaging

UN Sustainable Development Goals (SDGs)

This output has contributed to the advancement of the following goals:

Source: InCites