Article Text

Download PDFPDF
Automated segmentation of ultra-widefield fluorescein angiography of diabetic retinopathy using deep learning
  1. Phil-Kyu Lee1,
  2. Ho Ra1,
  3. Jiwon Baek1,2
  1. 1Department of Ophthalmology, Bucheon St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Bucheon, Republic of Korea
  2. 2Department of Ophthalmology, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea
  1. Correspondence to Dr Jiwon Baek, Department of Ophthalmology, The Catholic University of Korea, Seoul 02841, Korea (the Republic of); md.jiwon{at}gmail.com

Abstract

Background/Aims Retinal capillary non-perfusion (NP) and neovascularisation (NV) are two of the most important angiographic changes in diabetic retinopathy (DR). This study investigated the feasibility of using deep learning (DL) models to automatically segment NP and NV on ultra-widefield fluorescein angiography (UWFA) images from patients with DR.

Methods Retrospective cross-sectional chart review study. In total, 951 UWFA images were collected from patients with severe non-proliferative DR (NPDR) or proliferative DR (PDR). Each image was segmented and labelled for NP, NV, disc, background and outside areas. Using the labelled images, DL models were trained and validated (80%) using convolutional neural networks (CNNs) for automated segmentation and tested (20%) on test sets. Accuracy of each model and each label were assessed.

Results The best accuracy from CNN models for each label was 0.8208, 0.8338, 0.9801, 0.9253 and 0.9766 for NP, NV, disc, background and outside areas, respectively. The best Intersection over Union for each label was 0.6806, 0.5675, 0.7107, 0.8551 and 0.924 and mean mean boundary F1 score (BF score) was 0.6702, 0.8742, 0.9092, 0.8103 and 0.9006, respectively.

Conclusions DL models can detect NV and NP as well as disc and outer margins on UWFA with good performance. This automated segmentation of important UWFA features will aid physicians in DR clinics and in overcoming grader subjectivity.

  • retina
  • diagnostic tests/investigation
  • imaging

Data availability statement

Data are available on reasonable request.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Data availability statement

Data are available on reasonable request.

View Full Text

Footnotes

  • P-KL and HR contributed equally.

  • Contributors Conception and design of the work: JB. Acquisition, analysis or interpretation of data for the work: P-KL, HR and JB. Drafting the work: P-KL and JB. Revising and final approval of the version to be published: P-KL, HR and JB. Guarantor: JB.

  • Funding This work was supported by the Kim Ki-Soo Scholarship Committee (No. 2021) and National Research Foundation of Korea (NRF001663041G0003101).

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.