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

Traumatic intravitreal egg-shaped hematoma.

Eye (London, England)·2026
Same author

Bacterial keratitis: a global review of current practices, challenges and innovations.

The British journal of ophthalmology·2026
Same author

From topography to connectome: Towards an integrated understanding of the resting brain.

bioRxiv : the preprint server for biology·2026
Same author

Analysis of Corneal Axons Remodeling In Vivo during Development and Aging Using Long-Term Imaging in Mice.

The Journal of neuroscience : the official journal of the Society for Neuroscience·2026
Same author

Centenarian patients with neovascular age-related macular degeneration.

Retina (Philadelphia, Pa.)·2026
Same author

Corvis<sup>ST</sup> biomechanical indices in the diagnosis of corneal stromal and endothelial disorders: an artificial intelligence-based comparative study.

The British journal of ophthalmology·2025

Related Experiment Video

Updated: Jun 8, 2025

Three Different Protocols of Corneal Collagen Crosslinking in Keratoconus: Conventional, Accelerated and Iontophoresis
07:29

Three Different Protocols of Corneal Collagen Crosslinking in Keratoconus: Conventional, Accelerated and Iontophoresis

Published on: November 12, 2015

19.8K

Artificial intelligence versus conventional methods for RGP lens fitting in keratoconus.

Jérémy Abadou1, Simon Dahan2, Juliette Knoeri1

  • 1Groupe de Recherche Clinique #32, Transplantation et Thérapies Innovantes de la Cornée, Sorbonne Université, Hôpital National des 15-20, Paris, France.

Contact Lens & Anterior Eye : the Journal of the British Contact Lens Association
|November 5, 2024
PubMed
Summary

Artificial intelligence, especially Convolutional Neural Networks (CNNs), significantly improves the prediction of rigid contact lens posterior curvature in keratoconus eyes compared to traditional methods. This advancement offers better contact lens fitting for patients.

Keywords:
Contact lens fittingCorneal topographyDeep learningKeratoconusOptical coherence tomographyRigid gas permeable lens

More Related Videos

Full-Field Optical Coherence Microscopy for Histology-Like Analysis of Stromal Features in Corneal Grafts
07:51

Full-Field Optical Coherence Microscopy for Histology-Like Analysis of Stromal Features in Corneal Grafts

Published on: October 21, 2022

1.6K
Binocular Dynamic Visual Acuity in Eyeglass-Corrected Myopic Patients
07:06

Binocular Dynamic Visual Acuity in Eyeglass-Corrected Myopic Patients

Published on: March 29, 2022

2.5K

Related Experiment Videos

Last Updated: Jun 8, 2025

Three Different Protocols of Corneal Collagen Crosslinking in Keratoconus: Conventional, Accelerated and Iontophoresis
07:29

Three Different Protocols of Corneal Collagen Crosslinking in Keratoconus: Conventional, Accelerated and Iontophoresis

Published on: November 12, 2015

19.8K
Full-Field Optical Coherence Microscopy for Histology-Like Analysis of Stromal Features in Corneal Grafts
07:51

Full-Field Optical Coherence Microscopy for Histology-Like Analysis of Stromal Features in Corneal Grafts

Published on: October 21, 2022

1.6K
Binocular Dynamic Visual Acuity in Eyeglass-Corrected Myopic Patients
07:06

Binocular Dynamic Visual Acuity in Eyeglass-Corrected Myopic Patients

Published on: March 29, 2022

2.5K

Area of Science:

  • Ophthalmology
  • Biomedical Engineering
  • Artificial Intelligence

Background:

  • Keratoconus is a progressive eye condition affecting corneal shape.
  • Accurate prediction of the posterior radius of curvature of the best-fitted rigid contact lens (RCBFL) is crucial for successful contact lens fitting in keratoconus patients.
  • Current methods for RCBFL prediction have limitations in accuracy.

Purpose of the Study:

  • To compare the predictive efficiency of three artificial intelligence (AI) frameworks: Standard Machine Learning (ML), Multi-Layer Perceptron (MLP), and Convolutional Neural Networks (CNNs).
  • To evaluate these AI frameworks against a reference method (Mean radius of curvature, K) for predicting RCBFL in keratoconus eyes.
  • To determine the superior AI approach for RCBFL prediction using corneal topography data.

Main Methods:

  • A retrospective study of 197 keratoconus eyes from 135 patients fitted with Rose K2® rigid contact lenses.
  • Utilized topographic data (indices and reconstructed maps) from MS39® topographer for AI analysis.
  • Compared Standard ML, MLP (using indices), and CNNs (using maps) against the mean-K reference method, assessing accuracy with R-squared (r²) values.

Main Results:

  • Standard ML (Random Forest), MLP, and CNNs demonstrated significantly better RCBFL prediction than the mean-K reference (r²=0.36).
  • CNNs, particularly EfficientNetB0 trained with three topographic maps, achieved the highest prediction accuracy (r²=0.80).
  • AI methods showed superior performance, with accuracies ranging from 0.69 to 0.80 (p < 0.05).

Conclusions:

  • Artificial intelligence methods, especially CNNs, significantly outperform conventional approaches in predicting the posterior radius of curvature for rigid contact lenses in keratoconus.
  • Corneal topography data from the MS39® topographer is highly effective for AI-driven RCBFL prediction.
  • These findings suggest AI can enhance the precision of contact lens fitting for keratoconus patients.