Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Generalization of AI-Based Gestational Age Assessment Using Blind Sweep Ultrasonography.

JAMA network open·2026
Same author

Timing of Food Introduction and Allergy Prevention: An Update.

Immunology and allergy clinics of North America·2026
Same author

Biallelic Loss-Of-Function Variant in ATP5ME Is Associated With Severe and Early Onset Oxidative Phosphorylation Deficiency.

Journal of inherited metabolic disease·2026
Same author

Single-round modified Delphi consensus to optimise the care of patients with sickle cell disease across multidisciplinary teams within the NHS.

BMJ open·2026
Same author

Risk of cardiovascular events and death following retinal arterial occlusion and transient monocular visual loss.

Scientific reports·2026
Same author

Towards Conversational AI for Disease Management.

Nature·2026

Related Experiment Video

Updated: Oct 7, 2025

Quantification of Diabetes-induced Adherent Leukocytes in Retinal Vasculature
05:54

Quantification of Diabetes-induced Adherent Leukocytes in Retinal Vasculature

Published on: January 24, 2025

424

Deep Learning to Detect OCT-derived Diabetic Macular Edema from Color Retinal Photographs: A Multicenter Validation

Xinle Liu1, Tayyeba K Ali2, Preeti Singh1

  • 1Google Health, Google LLC, Mountain View, California.

Ophthalmology. Retina
|January 9, 2022
PubMed
Summary
This summary is machine-generated.

A deep learning system (DLS) accurately detects diabetic macular edema (DME) from fundus photos, outperforming human experts in specificity and sensitivity across diverse international populations. This technology could reduce unnecessary referrals in diabetic retinopathy screening.

Keywords:
artificial intelligencedeep learningdiabetic macular edemadiabetic retinopathyoptical coherence tomography

More Related Videos

Author Spotlight: Ex Vivo OCT-Based Multimodal Imaging of Human Donor Eyes for Research into Age-Related Macular Degeneration
10:14

Author Spotlight: Ex Vivo OCT-Based Multimodal Imaging of Human Donor Eyes for Research into Age-Related Macular Degeneration

Published on: May 26, 2023

3.7K
Using Retinal Imaging to Study Dementia
09:17

Using Retinal Imaging to Study Dementia

Published on: November 6, 2017

21.8K

Related Experiment Videos

Last Updated: Oct 7, 2025

Quantification of Diabetes-induced Adherent Leukocytes in Retinal Vasculature
05:54

Quantification of Diabetes-induced Adherent Leukocytes in Retinal Vasculature

Published on: January 24, 2025

424
Author Spotlight: Ex Vivo OCT-Based Multimodal Imaging of Human Donor Eyes for Research into Age-Related Macular Degeneration
10:14

Author Spotlight: Ex Vivo OCT-Based Multimodal Imaging of Human Donor Eyes for Research into Age-Related Macular Degeneration

Published on: May 26, 2023

3.7K
Using Retinal Imaging to Study Dementia
09:17

Using Retinal Imaging to Study Dementia

Published on: November 6, 2017

21.8K

Area of Science:

  • Ophthalmology
  • Artificial Intelligence
  • Medical Imaging

Background:

  • Diabetic macular edema (DME) is a leading cause of vision loss in diabetic patients.
  • Accurate detection of DME is crucial for timely treatment and prevention of blindness.
  • Current diagnostic methods rely on expert interpretation of retinal imaging, which can be time-consuming and subject to variability.

Purpose of the Study:

  • To validate the generalizability of a deep learning system (DLS) for detecting diabetic macular edema (DME).
  • To assess the DLS's performance in detecting DME from 2-dimensional color fundus photographs (CFP) using 3-dimensional OCT as the reference standard.
  • To compare the DLS's diagnostic accuracy against expert graders across international datasets.

Main Methods:

  • Retrospective validation of a DLS using paired CFP and OCT images from diverse international screening populations.
  • The DLS was trained on DME labels derived from OCT, identifying retinal thickening or intraretinal fluid.
  • Performance was evaluated by comparing DLS outputs to expert maculopathy grades and a previous DLS version.

Main Results:

  • The DLS demonstrated 80% specificity and 81% sensitivity in a combined Australian, Indian, and Thai dataset, outperforming expert graders (59% specificity, 70% sensitivity).
  • The DLS showed significantly higher specificity (P = 0.008) and non-inferior sensitivity (P < 0.001) compared to experts.
  • In a UK dataset, the DLS achieved 80% specificity and 100% sensitivity, exceeding expert performance.

Conclusions:

  • The DLS exhibits strong generalizability across multiple international populations for DME detection.
  • The DLS's performance surpasses that of human experts in terms of specificity and sensitivity.
  • The DLS holds potential clinical value in reducing false-positive referrals, thereby easing the burden on specialist eye care services.