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

Risk-stratified clinical management of primary colorectal mucosa-associated lymphoid tissue lymphoma based on endoscopic morphology.

Scandinavian journal of gastroenterology·2026
Same author

What affects natural killer cell activity: a cross-sectional study.

Frontiers in immunology·2026
Same author

Effects of Psychological Interventions for Mental Health in Police Officers: A Systematic Review and Meta-Analysis.

Healthcare (Basel, Switzerland)·2026
Same author

Association of polysomnography findings of obstructive sleep apnea with non-arteritic anterior ischemic optic neuropathy manifestations.

Sleep & breathing = Schlaf & Atmung·2026
Same author

Association Between Complete Blood Cell Count (CBC)-Derived Inflammatory Markers and Natural Killer Cell Activity: A Cross-Sectional Study.

Journal of inflammation research·2026
Same author

Defining Metal-Impurity Thresholds for Hydrogen Evolution in Sealed Vanadium Ion Batteries.

ACS omega·2026

Related Experiment Video

Updated: Jul 6, 2025

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

2.8K

Deep learning-based optic disc classification is affected by optic-disc tilt.

Youngwoo Nam1,2, Joonhyoung Kim3, Kyunga Kim2,4,5

  • 1Medical AI Research Center, Institute of Smart Healthcare, Samsung Medical Center, Seoul, Republic of Korea.

Scientific Reports
|January 4, 2024
PubMed
Summary
This summary is machine-generated.

Optic disc tilt significantly impacts deep learning classification accuracy for optic disc appearance. Models trained on non-tilted discs perform better, highlighting the need to account for tilt in AI development.

More Related Videos

In vivo Structural Assessments of Ocular Disease in Rodent Models using Optical Coherence Tomography
07:44

In vivo Structural Assessments of Ocular Disease in Rodent Models using Optical Coherence Tomography

Published on: July 24, 2020

3.0K
Using Retinal Imaging to Study Dementia
09:17

Using Retinal Imaging to Study Dementia

Published on: November 6, 2017

21.6K

Related Experiment Videos

Last Updated: Jul 6, 2025

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

2.8K
In vivo Structural Assessments of Ocular Disease in Rodent Models using Optical Coherence Tomography
07:44

In vivo Structural Assessments of Ocular Disease in Rodent Models using Optical Coherence Tomography

Published on: July 24, 2020

3.0K
Using Retinal Imaging to Study Dementia
09:17

Using Retinal Imaging to Study Dementia

Published on: November 6, 2017

21.6K

Area of Science:

  • Ophthalmology
  • Artificial Intelligence
  • Medical Imaging

Background:

  • Deep learning models are increasingly used for optic disc analysis in fundus images.
  • Optic disc tilt is a common anatomical variation that may affect classification performance.

Purpose of the Study:

  • To evaluate the influence of optic disc tilt on the accuracy of deep learning-based optic disc classification.
  • To compare the performance of deep learning models trained on tilted versus non-tilted optic disc datasets.

Main Methods:

  • Utilized a dataset of 2507 fundus photographs from 1809 subjects, with 40.3% exhibiting tilted optic discs.
  • Annotated images for normal and pathological optic disc changes (glaucoma, swelling, pallor).
  • Developed and compared deep learning classification models (VGG16, VGG19, DenseNet121) using all subjects, and separately for tilted and non-tilted disc groups.

Main Results:

  • Classification models demonstrated superior performance (higher AUC) when trained on non-tilted discs compared to tilted discs across all tested algorithms.
  • Models developed using non-tilted disc data achieved better sensitivity and specificity for classifying optic disc pathologies.
  • Models trained on the entire dataset showed reduced accuracy for subjects with tilted optic discs.

Conclusions:

  • Optic disc tilt is a critical factor that negatively affects the performance of deep learning-based optic disc classification.
  • Future development of optic disc classification algorithms requires methods to identify and adjust for the effects of optic disc tilt.