Automatic Retinal and Choroidal Boundary Segmentation in OCT Images Using Patch-Based Supervised Machine Learning Methods

David Alonso-Caneiro, Jason Kugelman, Jared Hamwood, Scott A. Read, Stephen J. Vincent, Fred K. Chen, Michael J. Collins

Research output: Chapter in Book/Conference paperConference paper

Abstract

The assessment of retinal and choroidal thickness derived from spectral domain optical coherence tomography (SD-OCT) images is an important clinical and research task. Current OCT instruments allow the capture of densely sampled, high-resolution cross-sectional images of ocular tissues. The extensive nature of such datasets makes the manual delineation of tissue boundaries time-consuming and impractical, especially for large datasets of images. Therefore, the development of reliable and accurate methods to automatically segment tissue boundaries in OCT images is fundamental. In this work, two different deep learning methods; convolutional neural networks (CNN) and recurrent neural networks (RNN) are evaluated to calculate the probability of the retinal and choroidal boundaries of interest to be located in a specific position within the SD-OCT images. The method is initially trained using small image patches centred around the three boundaries of interest. After that, the method can be used to provide a per-layer probability map that marks the most likely location of the boundaries. To convert each layer-probability map into a boundary position, the map is subsequently traced using a graph-search method. The effect of the network architecture (CNN vs RNN), patch size, and image intensity compensation on the performance and subsequent boundary segmentation is presented. The results are compared with manual boundary segmentation as well as a previously proposed method based on standard image analysis techniques.

Original languageEnglish
Title of host publicationComputer Vision – ACCV 2018 Workshops - 14th Asian Conference on Computer Vision, 2018, Revised Selected Papers
EditorsGustavo Carneiro, Shaodi You
Place of PublicationPerth, Australia
PublisherSpringer-Verlag Berlin
Pages215-228
Number of pages14
ISBN (Print)9783030210731
DOIs
Publication statusPublished - 19 Jun 2019
Event14th Asian Conference on Computer Vision, ACCV 2018 - Perth, Australia
Duration: 2 Dec 20186 Dec 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11367 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference14th Asian Conference on Computer Vision, ACCV 2018
CountryAustralia
CityPerth
Period2/12/186/12/18

Fingerprint

Supervised Learning
Patch
Learning systems
Machine Learning
Segmentation
Recurrent neural networks
Optical tomography
Tissue
Neural networks
Network architecture
Optical Coherence Tomography
Image analysis
Recurrent Neural Networks
Neural Networks
Graph Search
Network Architecture
Search Methods
Image Analysis
Large Data Sets
Convert

Cite this

Alonso-Caneiro, D., Kugelman, J., Hamwood, J., Read, S. A., Vincent, S. J., Chen, F. K., & Collins, M. J. (2019). Automatic Retinal and Choroidal Boundary Segmentation in OCT Images Using Patch-Based Supervised Machine Learning Methods. In G. Carneiro, & S. You (Eds.), Computer Vision – ACCV 2018 Workshops - 14th Asian Conference on Computer Vision, 2018, Revised Selected Papers (pp. 215-228). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11367 LNCS). Perth, Australia: Springer-Verlag Berlin. https://doi.org/10.1007/978-3-030-21074-8_17
Alonso-Caneiro, David ; Kugelman, Jason ; Hamwood, Jared ; Read, Scott A. ; Vincent, Stephen J. ; Chen, Fred K. ; Collins, Michael J. / Automatic Retinal and Choroidal Boundary Segmentation in OCT Images Using Patch-Based Supervised Machine Learning Methods. Computer Vision – ACCV 2018 Workshops - 14th Asian Conference on Computer Vision, 2018, Revised Selected Papers. editor / Gustavo Carneiro ; Shaodi You. Perth, Australia : Springer-Verlag Berlin, 2019. pp. 215-228 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{256298fbfb0e4b35a6d8de418e3297f9,
title = "Automatic Retinal and Choroidal Boundary Segmentation in OCT Images Using Patch-Based Supervised Machine Learning Methods",
abstract = "The assessment of retinal and choroidal thickness derived from spectral domain optical coherence tomography (SD-OCT) images is an important clinical and research task. Current OCT instruments allow the capture of densely sampled, high-resolution cross-sectional images of ocular tissues. The extensive nature of such datasets makes the manual delineation of tissue boundaries time-consuming and impractical, especially for large datasets of images. Therefore, the development of reliable and accurate methods to automatically segment tissue boundaries in OCT images is fundamental. In this work, two different deep learning methods; convolutional neural networks (CNN) and recurrent neural networks (RNN) are evaluated to calculate the probability of the retinal and choroidal boundaries of interest to be located in a specific position within the SD-OCT images. The method is initially trained using small image patches centred around the three boundaries of interest. After that, the method can be used to provide a per-layer probability map that marks the most likely location of the boundaries. To convert each layer-probability map into a boundary position, the map is subsequently traced using a graph-search method. The effect of the network architecture (CNN vs RNN), patch size, and image intensity compensation on the performance and subsequent boundary segmentation is presented. The results are compared with manual boundary segmentation as well as a previously proposed method based on standard image analysis techniques.",
author = "David Alonso-Caneiro and Jason Kugelman and Jared Hamwood and Read, {Scott A.} and Vincent, {Stephen J.} and Chen, {Fred K.} and Collins, {Michael J.}",
year = "2019",
month = "6",
day = "19",
doi = "10.1007/978-3-030-21074-8_17",
language = "English",
isbn = "9783030210731",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer-Verlag Berlin",
pages = "215--228",
editor = "Gustavo Carneiro and Shaodi You",
booktitle = "Computer Vision – ACCV 2018 Workshops - 14th Asian Conference on Computer Vision, 2018, Revised Selected Papers",

}

Alonso-Caneiro, D, Kugelman, J, Hamwood, J, Read, SA, Vincent, SJ, Chen, FK & Collins, MJ 2019, Automatic Retinal and Choroidal Boundary Segmentation in OCT Images Using Patch-Based Supervised Machine Learning Methods. in G Carneiro & S You (eds), Computer Vision – ACCV 2018 Workshops - 14th Asian Conference on Computer Vision, 2018, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11367 LNCS, Springer-Verlag Berlin, Perth, Australia, pp. 215-228, 14th Asian Conference on Computer Vision, ACCV 2018, Perth, Australia, 2/12/18. https://doi.org/10.1007/978-3-030-21074-8_17

Automatic Retinal and Choroidal Boundary Segmentation in OCT Images Using Patch-Based Supervised Machine Learning Methods. / Alonso-Caneiro, David; Kugelman, Jason; Hamwood, Jared; Read, Scott A.; Vincent, Stephen J.; Chen, Fred K.; Collins, Michael J.

Computer Vision – ACCV 2018 Workshops - 14th Asian Conference on Computer Vision, 2018, Revised Selected Papers. ed. / Gustavo Carneiro; Shaodi You. Perth, Australia : Springer-Verlag Berlin, 2019. p. 215-228 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11367 LNCS).

Research output: Chapter in Book/Conference paperConference paper

TY - GEN

T1 - Automatic Retinal and Choroidal Boundary Segmentation in OCT Images Using Patch-Based Supervised Machine Learning Methods

AU - Alonso-Caneiro, David

AU - Kugelman, Jason

AU - Hamwood, Jared

AU - Read, Scott A.

AU - Vincent, Stephen J.

AU - Chen, Fred K.

AU - Collins, Michael J.

PY - 2019/6/19

Y1 - 2019/6/19

N2 - The assessment of retinal and choroidal thickness derived from spectral domain optical coherence tomography (SD-OCT) images is an important clinical and research task. Current OCT instruments allow the capture of densely sampled, high-resolution cross-sectional images of ocular tissues. The extensive nature of such datasets makes the manual delineation of tissue boundaries time-consuming and impractical, especially for large datasets of images. Therefore, the development of reliable and accurate methods to automatically segment tissue boundaries in OCT images is fundamental. In this work, two different deep learning methods; convolutional neural networks (CNN) and recurrent neural networks (RNN) are evaluated to calculate the probability of the retinal and choroidal boundaries of interest to be located in a specific position within the SD-OCT images. The method is initially trained using small image patches centred around the three boundaries of interest. After that, the method can be used to provide a per-layer probability map that marks the most likely location of the boundaries. To convert each layer-probability map into a boundary position, the map is subsequently traced using a graph-search method. The effect of the network architecture (CNN vs RNN), patch size, and image intensity compensation on the performance and subsequent boundary segmentation is presented. The results are compared with manual boundary segmentation as well as a previously proposed method based on standard image analysis techniques.

AB - The assessment of retinal and choroidal thickness derived from spectral domain optical coherence tomography (SD-OCT) images is an important clinical and research task. Current OCT instruments allow the capture of densely sampled, high-resolution cross-sectional images of ocular tissues. The extensive nature of such datasets makes the manual delineation of tissue boundaries time-consuming and impractical, especially for large datasets of images. Therefore, the development of reliable and accurate methods to automatically segment tissue boundaries in OCT images is fundamental. In this work, two different deep learning methods; convolutional neural networks (CNN) and recurrent neural networks (RNN) are evaluated to calculate the probability of the retinal and choroidal boundaries of interest to be located in a specific position within the SD-OCT images. The method is initially trained using small image patches centred around the three boundaries of interest. After that, the method can be used to provide a per-layer probability map that marks the most likely location of the boundaries. To convert each layer-probability map into a boundary position, the map is subsequently traced using a graph-search method. The effect of the network architecture (CNN vs RNN), patch size, and image intensity compensation on the performance and subsequent boundary segmentation is presented. The results are compared with manual boundary segmentation as well as a previously proposed method based on standard image analysis techniques.

UR - http://www.scopus.com/inward/record.url?scp=85068458758&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-21074-8_17

DO - 10.1007/978-3-030-21074-8_17

M3 - Conference paper

SN - 9783030210731

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 215

EP - 228

BT - Computer Vision – ACCV 2018 Workshops - 14th Asian Conference on Computer Vision, 2018, Revised Selected Papers

A2 - Carneiro, Gustavo

A2 - You, Shaodi

PB - Springer-Verlag Berlin

CY - Perth, Australia

ER -

Alonso-Caneiro D, Kugelman J, Hamwood J, Read SA, Vincent SJ, Chen FK et al. Automatic Retinal and Choroidal Boundary Segmentation in OCT Images Using Patch-Based Supervised Machine Learning Methods. In Carneiro G, You S, editors, Computer Vision – ACCV 2018 Workshops - 14th Asian Conference on Computer Vision, 2018, Revised Selected Papers. Perth, Australia: Springer-Verlag Berlin. 2019. p. 215-228. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-21074-8_17