This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies.  Read our Cookies Policy.
Close
Eye News
  • Features
    • Close
    • Features
    • Allied Professions
    • Humanitarian
    • Interviews
    • AI & Oculomics
    • Ophthalmology
    • Optometry
    • Podcast videos
    • Supplements
  • Education
    • Close
    • Education
    • Learning Curve
    • Quiz
    • Top Tips
    • Trainees
    • Medico-Legal
    • The Truth Behind The Headlines
    • Case Reports
    • Pete's Bogus Journey
  • Reviews
    • Close
    • Reviews
    • Book Reviews
    • Journal Reviews
    • What's trending?
    • Tech Reviews
    • My Top Five
    • The Culture Section
  • Events
  • News
  • Product Guide
  • Industry News
  • Contact us
    • Close
    • Contact us
    • Write for Eye News
  • Home
  • Reviews
  • Journal Reviews
  • Automated grading of diabetic retinopathy: is it possible?

Automated grading of diabetic retinopathy: is it possible?
Reviewed by Eulee Seow

1 August 2018 | Eulee Seow | EYE - Vitreo-Retinal

Diabetic retinopathy (DR) is a frequent microvascular complication of diabetes and a leading cause of blindness worldwide. However, much of this diabetic blindness can be delayed or even prevented with timely diagnosis and proper treatment. For this reason, regular screening and early detection of potentially sight-threatening retinopathy is a key strategy. Currently in the UK trained graders are used for screening. This study looks at the sensitivity and specificity of an automated algorithm for detecting referral-warranted DR on Optos ultrawidefield (UWF) pseudocolour images. A total of 383 subjects (754 eyes) were enrolled. Nonproliferative DR graded to be moderate or higher on the 5-level International Clinical Diabetic Retinopathy (ICDR) severity scale was considered as grounds for referral. The software automatically detected DR lesions using the previously trained classifiers and classified each image in the test set as referral-warranted or not warranted. Sensitivity, specificity and the area under the receiver operating curve (AUROC) of the algorithm were computed. The automated algorithm achieved a 91.7%/90.3% sensitivity (95% CI 90.1–93.9/80.4–89.4) with a 50.0%/53.6% specificity (95% CI 31.7–72.8/36.5–71.4) for detecting referral-warranted retinopathy at the patient / eye levels, respectively; the AUROC was 0.873/0.851 (95% CI 0.819–0.922/0.804–0.894). They concluded that DR lesions were detected from Optos pseudocolour UWF images using an automated algorithm. Images were classified as referral-warranted DR with a high degree of sensitivity and moderate specificity. Automated analysis of UWF images could be of value in DR screening programmes and could allow for more complete and accurate disease staging.

Automated detection of diabetic retinopathy lesions on ultrawidefield pseudocolour images.
Wang K, Jayadev C, Nittala M, et al.
ACTA OPHTHALMOLOGICA
2018;96:e168-73.
Share This
CONTRIBUTOR
Eulee Seow

University Hospital of Wales, UK.

View Full Profile
Specialty
  • EYE - Cataract
  • EYE - Cornea
  • EYE - General
  • EYE - Glaucoma
  • EYE - Neuro-ophthalmology
  • EYE - Oculoplastic
  • EYE - Oncology
  • EYE - Orbit
  • EYE - Paediatrics
  • EYE - Pathology
  • EYE - Refractive
  • EYE - Strabismus
  • EYE - Vitreo-Retinal
Archive
  • 2025
  • 2024
  • 2023
  • 2022
  • 2021
  • 2020
  • 2019
  • 2018
  • 2017
  • 2016
  • 2015
  • 2014
  • 2013

Top Of Page

9 Gayfield Square, 
Edinburgh EH1 3NT, UK.

Call: +44 (0)131 557 4184
www.pinpoint-scotland.com

WEBSITE DETAILS
  • Cookie Policy
  • Data Protection Notice
  • Privacy Policy
  • Terms and Conditions
ABOUT US
  • Who we are
  • Register
  • Contact us
  • Contributors
  • Company Awards
DIGITAL ISSUES/GUIDELINES
  • Digital issues - Library
  • Supplements - Library
  • Guidelines
Accreditations
IPSO_FLAG_TEAL 2025.png cpdcertified.png

Pinpoint Scotland Ltd (Registered in Scotland No. SC068684) | © 2025 - Website by Gecko Agency