Journal article
A Deep-Learning Approach for Automated OCT En-Face Retinal Vessel Segmentation in Cases of Optic Disc Swelling Using Multiple En-Face Images as Input
Translational vision science & technology, Vol.9(2), pp.17-17
03/2020
DOI: 10.1167/tvst.9.2.17
PMCID: PMC7401896
PMID: 32821471
Abstract
In cases of optic disc swelling, segmentation of projected retinal blood vessels from optical coherence tomography (OCT) volumes is challenging due to swelling-based shadowing artifacts. Based on our hypothesis that simultaneously considering vessel information from multiple projected retinal layers can substantially increase vessel visibility, in this work, we propose a deep-learning-based approach to segment vessels involving the simultaneous use of three OCT en-face images as input.
A human expert vessel tracing combining information from OCT en-face images of the retinal pigment epithelium (RPE), inner retina, and total retina as well as a registered fundus image served as the reference standard. The deep neural network was trained from the imaging data from 18 patients with optic disc swelling to output a vessel probability map from three OCT en-face input images. The vessels from the OCT en-face images were also manually traced in three separate stages to compare with the performance of the proposed approach.
On an independent volume-matched test set of 18 patients, the proposed deep-learning-based approach outperformed the three OCT-based manual tracing stages. The manual tracing based on three OCT en-face images also outperformed the manual tracing using only the traditional RPE en-face image.
In cases of optic disc swelling, use of multiple en-face images enables better vessel segmentation when compared with the traditional use of a single en-face image.
Improved vessel segmentation approaches in cases of optic disc swelling can be used as features for an improved assessment of the severity and cause of the swelling.
Details
- Title: Subtitle
- A Deep-Learning Approach for Automated OCT En-Face Retinal Vessel Segmentation in Cases of Optic Disc Swelling Using Multiple En-Face Images as Input
- Creators
- Mohammad Shafkat Islam - Department of Electrical and Computer Engineering, University of Iowa, Iowa City, IA, USAJui-Kai Wang - Iowa City VA Health Care System and Iowa City VA Center for the Prevention and Treatment of Visual Loss, Iowa City, IA, USASamuel S Johnson - Department of Electrical and Computer Engineering, University of Iowa, Iowa City, IA, USAMatthew J Thurtell - Department of Ophthalmology and Visual Sciences, University of Iowa, Iowa City, IA, USARandy H Kardon - Department of Ophthalmology and Visual Sciences, University of Iowa, Iowa City, IA, USAMona K Garvin - Iowa City VA Health Care System and Iowa City VA Center for the Prevention and Treatment of Visual Loss, Iowa City, IA, USA
- Resource Type
- Journal article
- Publication Details
- Translational vision science & technology, Vol.9(2), pp.17-17
- Publisher
- United States
- DOI
- 10.1167/tvst.9.2.17
- PMID
- 32821471
- PMCID
- PMC7401896
- ISSN
- 2164-2591
- eISSN
- 2164-2591
- Grant note
- R01 EY023279 / NEI NIH HHS I01 RX001786 / RRD VA I50 RX003002 / RRD VA
- Language
- English
- Date published
- 03/2020
- Academic Unit
- Neurology; Electrical and Computer Engineering; Iowa Neuroscience Institute; Ophthalmology and Visual Sciences
- Record Identifier
- 9984070883402771
Metrics
23 Record Views