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

Clinically grounded retinal representation learning from minimal supervision.

Scientific reports·2026
Same author

Fast progressors in Asian normal-tension glaucoma: 10 years and beyond in a longitudinal cohort.

The British journal of ophthalmology·2026
Same author

Retinal Arteriosclerosis in a Large Health Screening Cohort: Associations with Systemic Vascular Comorbidities and Stroke in Young Adults.

Biomedicines·2026
Same author

Interpreting the structure-function relationship in high myopia-associated glaucoma: A review.

Acta ophthalmologica·2026
Same author

Timing of Moderate-to-Vigorous Physical Activity and Mortality Risk: Role of Chronotype and Sleep Patterns.

European journal of preventive cardiology·2026
Same author

A comparative analysis of pig-to-rhesus corneal xenotransplantation with various immunosuppressive regimens.

Korean journal of ophthalmology : KJO·2026

Related Experiment Video

Updated: Nov 27, 2025

Assessing Binocular Central Visual Field and Binocular Eye Movements in a Dichoptic Viewing Condition
07:45

Assessing Binocular Central Visual Field and Binocular Eye Movements in a Dichoptic Viewing Condition

Published on: July 21, 2020

4.7K

Estimating visual field loss from monoscopic optic disc photography using deep learning model.

Jinho Lee1,2, Yong Woo Kim1,3, Ahnul Ha1,4

  • 1Department of Ophthalmology, Seoul National University College of Medicine, Seoul, Korea.

Scientific Reports
|December 4, 2020
PubMed
Summary

A novel deep learning algorithm can predict glaucoma

More Related Videos

Using Retinal Imaging to Study Dementia
09:17

Using Retinal Imaging to Study Dementia

Published on: November 6, 2017

22.0K
Assessing Early Stage Open-Angle Glaucoma in Patients by Isolated-Check Visual Evoked Potential
07:11

Assessing Early Stage Open-Angle Glaucoma in Patients by Isolated-Check Visual Evoked Potential

Published on: May 25, 2020

6.6K

Related Experiment Videos

Last Updated: Nov 27, 2025

Assessing Binocular Central Visual Field and Binocular Eye Movements in a Dichoptic Viewing Condition
07:45

Assessing Binocular Central Visual Field and Binocular Eye Movements in a Dichoptic Viewing Condition

Published on: July 21, 2020

4.7K
Using Retinal Imaging to Study Dementia
09:17

Using Retinal Imaging to Study Dementia

Published on: November 6, 2017

22.0K
Assessing Early Stage Open-Angle Glaucoma in Patients by Isolated-Check Visual Evoked Potential
07:11

Assessing Early Stage Open-Angle Glaucoma in Patients by Isolated-Check Visual Evoked Potential

Published on: May 25, 2020

6.6K

Area of Science:

  • Ophthalmology
  • Medical Imaging
  • Artificial Intelligence

Background:

  • Visual field assessment is crucial for diagnosing glaucoma but suffers from test-retest variability.
  • Optic disc photographs (ODPs) offer a potential alternative imaging source for assessing glaucomatous damage.
  • Standard automated perimetry (SAP) results can be variable, necessitating complementary diagnostic methods.

Purpose of the Study:

  • To develop and evaluate a deep learning (DL) algorithm for predicting mean deviation (MD) in standard automated perimetry (SAP) using monoscopic optic disc photographs (ODPs).
  • To assess the DL algorithm's ability to detect glaucomatous visual field (VF) loss.
  • To establish the feasibility of using ODPs for quantitative glaucoma assessment via DL.

Main Methods:

  • A deep learning model was constructed by integrating a pre-trained network with fully connected layers.
  • The model was trained and validated using 1000 image pairs (ODPs and SAP results) from 563 eyes.
  • Performance was evaluated on a separate test set of 200 image pairs, calculating correlation coefficient, mean absolute error (MAE), and area under the receiver operating characteristic curve (AUC).

Main Results:

  • The DL algorithm demonstrated a strong correlation (coefficient=0.755) and good agreement (R²=57.0%, MAE=1.94 dB) between predicted and actual MD values.
  • The model achieved high accuracy in detecting glaucomatous VF loss, with an AUC of 0.953.
  • The DL algorithm showed significant feasibility in predicting MD and identifying functional loss from ODPs.

Conclusions:

  • Deep learning algorithms can accurately predict visual field parameters from optic disc photographs.
  • This DL approach offers a promising, less variable method for detecting glaucomatous functional loss.
  • Optic disc photographs analyzed by DL present a viable tool for glaucoma assessment, potentially overcoming limitations of traditional visual field testing.