Isfahan University of Medical Sciences

Science Communicator Platform

Stay connected! Follow us on X network (Twitter):
Share this content! On (X network) By
Automatic Choroid Vascularity Index Calculation in Optical Coherence Tomography Images With Low-Contrast Sclerochoroidal Junction Using Deep Learning Publisher



Arian R1 ; Mahmoudi T2 ; Riaziesfahani H3 ; Faghihi H3 ; Mirshahi A3 ; Ghassemi F3 ; Khodabande A3 ; Kafieh R1, 4 ; Khalili Pour E3
Authors
Show Affiliations
Authors Affiliations
  1. 1. Medical Image and Signal Processing Research Center, School of Advanced Technologies in Medicine, Isfahan University of Medical Sciences, Isfahan, 81746-73461, Iran
  2. 2. Department of Medical Physics and Biomedical Engineering, School of Medicine, Shiraz University of Medical Sciences, Shiraz, 71348-14336, Iran
  3. 3. Farabi Eye Hospital, Tehran University of Medical Sciences, Retina Ward, Tehran, 14176-13151, Iran
  4. 4. Department of Engineering, Durham University, South Road, Durham, DH1 3LE, United Kingdom

Source: Photonics Published:2023


Abstract

The choroidal vascularity index (CVI) is a new biomarker defined for retinal optical coherence tomography (OCT) images for measuring and evaluating the choroidal vascular structure. The CVI is the ratio of the choroidal luminal area (LA) to the total choroidal area (TCA). The automatic calculation of this index is important for ophthalmologists but has not yet been explored. In this study, we proposed a fully automated method based on deep learning for calculating the CVI in three main steps: 1—segmentation of the choroidal boundary, 2—detection of the choroidal luminal vessels, and 3—computation of the CVI. The proposed method was evaluated in complex situations such as the presence of diabetic retinopathy and pachychoroid spectrum. In pachychoroid spectrum, the choroid is thickened, and the boundary between the choroid and sclera (sclerochoroidal junction) is blurred, which makes the segmentation more challenging. The proposed method was designed based on the U-Net model, and a new loss function was proposed to overcome the segmentation problems. The vascular LA was then calculated using Niblack’s local thresholding method, and the CVI value was finally computed. The experimental results for the segmentation stage with the best-performing model and the proposed loss function used showed Dice coefficients of 0.941 and 0.936 in diabetic retinopathy and pachychoroid spectrum patients, respectively. The unsigned boundary localization errors in the presence of diabetic retinopathy were 3 and 20.7 μm for the BM boundary and sclerochoroidal junction, respectively. Similarly, the unsigned errors in the presence of pachychoroid spectrum were 21.6 and 76.2 μm for the BM and sclerochoroidal junction, respectively. The performance of the proposed method to calculate the CVI was evaluated; the Bland–Altman plot indicated an acceptable agreement between the values allocated by experts and the proposed method in the presence of diabetic retinopathy and pachychoroid spectrum. © 2023 by the authors.
Other Related Docs
9. Introduction to Optical Coherence Tomography, Atlas of Ocular Optical Coherence Tomograph, Second Edition (2023)
13. Determination of Foveal Avascular Zone Parameters Using a New Location-Aware Deep-Learning Method, Proceedings of SPIE - The International Society for Optical Engineering (2021)
16. A Computationally Efficient Red-Lesion Extraction Method for Retinal Fundus Images, IEEE Transactions on Instrumentation and Measurement (2023)
17. Exact Localization of Breakpoints of Retinal Pigment Epithelium in Optical Coherence Tomography of Optic Nerve Head, Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS (2017)
35. Detection of Retinal Abnormalities in Oct Images Using Wavelet Scattering Network, Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS (2022)
37. Analysis of Foveal Avascular Zone for Grading of Diabetic Retinopathy Severity Based on Curvelet Transform, Graefe's Archive for Clinical and Experimental Ophthalmology (2012)
39. Automatic Detection of Microaneurysms in Oct Images Using Bag of Features, Computational and Mathematical Methods in Medicine (2022)
44. Stochastic Differential Equations for Automatic Quality Control of Retinal Optical Coherence Tomography Images, Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS (2022)
48. Comparison of Macular Octs in Right and Left Eyes of Normal People, Progress in Biomedical Optics and Imaging - Proceedings of SPIE (2014)
49. Automatic Classification of Macular Diseases From Oct Images Using Cnn Guided With Edge Convolutional Layer, Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS (2022)