Segmentation of retinal fluid based on deep learning: application of three-dimensional fully convolutional neural networks in optical coherence tomography images

Journal Title: International Journal of Ophthalmology - Year 2019, Vol 12, Issue 6

Abstract

"AIM: To explore a segmentation algorithm based on deep learning to achieve accurate diagnosis and treatment of patients with retinal fluid. METHODS: A two-dimensional (2D) fully convolutional network for retinal segmentation was employed. In order to solve the category imbalance in retinal optical coherence tomography (OCT) images, the network parameters and loss function based on the 2D fully convolutional network were modified. For this network, the correlations of corresponding positions among adjacent images in space are ignored. Thus, we proposed a three-dimensional (3D) fully convolutional network for segmentation in the retinal OCT images. RESULTS: The algorithm was evaluated according to segmentation accuracy, Kappa coefficient, and F1 score. For the 3D fully convolutional network proposed in this paper, the overall segmentation accuracy rate is 99.56%, Kappa coefficient is 98.47%, and F1 score of retinal fluid is 95.50%. CONCLUSION: The OCT image segmentation algorithm based on deep learning is primarily founded on the 2D convolutional network. The 3D network architecture proposed in this paper reduces the influence of category imbalance, realizes end-to-end segmentation of volume images, and achieves optimal segmentation results. The segmentation maps are practically the same as the manual annotations of doctors, and can provide doctors with more accurate diagnostic data."

Authors and Affiliations

Tian-Wei Qian

Keywords

Related Articles

Hydrogen sulfide intervention in cystathionine-β-synthase mutant mouse helps restore ocular homeostasis

"AIM: To investigate the applications of hydrogen sulfide (H2S) in eye-specific ailments in mice. METHODS: Heterozygous cystathionine-β-synthase (CBS+/-) and wild-type C57BL/6J (WT) mice fed with or without high methionin...

Effects of chronic elevated intraocular pressure on parameters of optical coherence tomography in rhesus monkeys

"AIM: To determine the progression of parameters from optical coherence tomography (OCT) in chronic elevated intraocular pressure (IOP) monkeys. METHODS: A chronic elevated IOP model of rhesus monkeys was induced by lase...

Biomechanics of sclera crosslinked using genipin in rabbit

"AIM: To strengthen the biomechanics of collagen by crosslinking rabbit scleral collagen with genipin to develop a new therapy for preventing myopic progression. METHODS: Ten New Zealand rabbits were treated with 0.5 mmo...

A clinicopathological study on IgG4-related ophthalmic disease

"AIM: To investigate clinicopathological features of immunoglobulin G4-related ophthalmic disease (IgG4-ROD), and analyze the recurrence rates following systemic corticosteroid administration. METHODS: We retrospectively...

Ex-PRESS implantation versus trabeculectomy in Chinese patients with POAG: fellow eye pilot study

"AIM: To compare the outcomes of Ex-PRESS implantation in one eye versus trabeculectomy with mitomycin C in the fellow eye in Chinese patients with primary open-angle glaucoma (POAG). METHODS: This was a prospective, non...

Download PDF file
  • EP ID EP642905
  • DOI 10.18240/ijo.2019.06.22
  • Views 40
  • Downloads 0

How To Cite

Tian-Wei Qian (2019). Segmentation of retinal fluid based on deep learning: application of three-dimensional fully convolutional neural networks in optical coherence tomography images. International Journal of Ophthalmology, 12(6), 1012-1020. https://europub.co.uk./articles/-A-642905