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 Experiment Video

Updated: Jan 4, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.3K

Validation of a Deep Learning Algorithm for Diabetic Retinopathy.

Pedro Romero-Aroca1, Raquel Verges-Puig1, Jordi de la Torre2

  • 1Ophthalmic Department, University Hospital Sant Joan, Pere Virgili Institute (IISPV), University Rovira and Virgili, Reus, Spain.

Telemedicine Journal and E-Health : the Official Journal of the American Telemedicine Association
|November 5, 2019
PubMed
Summary
This summary is machine-generated.

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

Longitudinal evaluation of tumor-infiltrating lymphocyte scoring using automated region of interest registration in breast cancer.

Breast cancer research : BCR·2026
Same author

Automated Multi-Modal MRI Segmentation of Stroke Lesions and Corticospinal Tract Integrity for Functional Outcome Prediction.

Tomography (Ann Arbor, Mich.)·2026
Same author

DDFU-Net: A Deep Decoder-Focused U-Net Model for Retinal Lesion Segmentation.

Annals of biomedical engineering·2026
Same author

Scalable unit harmonization in medical informatics via Bayesian-optimized retrieval and transformer-based re-ranking.

International journal of medical informatics·2025
Same author

Radiogenomics for Glioblastoma Survival Prediction: Integrating Radiomics, Clinical, and Genomic Features Using Artificial Intelligence.

Journal of imaging informatics in medicine·2025
Same author

Advancing breast cancer relapse prediction with radiomics and neural networks: a clinically interpretable framework.

Frontiers in oncology·2025
Same journal

Patient Preferences for Virtual Versus In-Person Care in Inflammatory Bowel Disease: A Cross-Sectional Study.

Telemedicine journal and e-health : the official journal of the American Telemedicine Association·2026
Same journal

A Qualitative Retrospective Chart Review of Remote Communication for Rural Older Cancer Survivors.

Telemedicine journal and e-health : the official journal of the American Telemedicine Association·2026
Same journal

Characterization of Clinical Diagnoses at a Community Telehealth Platform in Uganda.

Telemedicine journal and e-health : the official journal of the American Telemedicine Association·2026
Same journal

<i>Retraction:</i> Revolutionizing Sports Rehabilitation: Unleashing the Power of Tele-Rehabilitation for Optimal Physiotherapy Results.

Telemedicine journal and e-health : the official journal of the American Telemedicine Association·2026
Same journal

Teleconsultation Competencies in Undergraduate Medical Education: A Critical Narrative Review and Evidence-Gap Analysis.

Telemedicine journal and e-health : the official journal of the American Telemedicine Association·2026
Same journal

Variation in Telehealth Use in a National Home Test-to-Treat Program for Acute Respiratory Infections.

Telemedicine journal and e-health : the official journal of the American Telemedicine Association·2026
See all related articles

A deep learning algorithm (DLA) effectively screens for diabetic retinopathy (DR) in retinal images. This validated DLA demonstrates high accuracy in detecting any DR and referable DR, aiding early diagnosis.

Area of Science:

  • Ophthalmology
  • Artificial Intelligence
  • Medical Imaging

Background:

  • Diabetic retinopathy (DR) is a growing concern due to the increasing incidence of diabetes mellitus.
  • Current DR detection relies on manual interpretation of retinographies, necessitating more efficient screening systems.
  • Automated systems are crucial for managing the high volume of diabetes patients requiring DR screening.

Purpose of the Study:

  • To validate a deep learning algorithm (DLA) for the accurate detection of diabetic retinopathy (DR) from retinal images (retinographies).
  • To assess the DLA's performance in identifying both any DR and referable DR cases.
  • To establish the DLA's utility as a diagnostic aid in DR screening programs.

Main Methods:

  • A DLA was developed and trained on a large dataset of 110,000+ retinal images from multiple sources.
Keywords:
convolutional neural networkdeep learningdiabetic retinopathyscreening of diabetic retinopathy

More Related Videos

Tear-Derived Exosomal miR-15a as New Diagnostic Tool for Diabetic Retinopathy
07:45

Tear-Derived Exosomal miR-15a as New Diagnostic Tool for Diabetic Retinopathy

Published on: December 30, 2025

54
Author Spotlight: Understanding Retinal Vessel Resilience and Disease Progression
04:36

Author Spotlight: Understanding Retinal Vessel Resilience and Disease Progression

Published on: January 12, 2024

1.5K

Related Experiment Videos

Last Updated: Jan 4, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.3K
Tear-Derived Exosomal miR-15a as New Diagnostic Tool for Diabetic Retinopathy
07:45

Tear-Derived Exosomal miR-15a as New Diagnostic Tool for Diabetic Retinopathy

Published on: December 30, 2025

54
Author Spotlight: Understanding Retinal Vessel Resilience and Disease Progression
04:36

Author Spotlight: Understanding Retinal Vessel Resilience and Disease Progression

Published on: January 12, 2024

1.5K
  • A validation cohort of 38,339 retinographies from 17,669 patients was used to evaluate the DLA.
  • The DLA's performance was compared against diagnoses made by four senior retina ophthalmologists, using metrics like Cohen's weighted Kappa (CWK), sensitivity, specificity, and predictive values.

    • Main Results:

    • For any DR detection, the DLA achieved a CWK of 0.886, sensitivity of 96.7%, specificity of 97.6%, PPV of 83.6%, and NPV of 99.6%.
    • For referable DR detection, the DLA reported a CWK of 0.809, sensitivity of 99.8%, specificity of 96.8%, PPV of 70.1%, and NPV of 92.8%.
    • The DLA demonstrated low Type I (0.024 for any-DR, 0.032 for referable-DR) and Type II (0.004 for any-DR, 0.001 for referable-DR) errors.

    Conclusions:

    • The validated deep learning algorithm is a high-confidence tool for assisting in diabetic retinopathy screening.
    • The DLA accurately identifies patients with any DR and those requiring referral, supporting clinical decision-making.
    • The DLA shows significant potential to aid ophthalmologists and other healthcare professionals in DR screening, particularly in challenging cases.