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

Epidemiology of eyelid disorders in an elderly population: Tehran Geriatric Eye Study.

BMC ophthalmology·2026
Same author

Evaluating corneal topographic patterns in keratoconic eyes and their association with keratoconus changes over time.

International journal of ophthalmology·2026
Same author

Statistical shape modeling as a practical tool for dosimetry: quantifying uncertainty in internal dose assessment.

Physica medica : PM : an international journal devoted to the applications of physics to medicine and biology : official journal of the Italian Association of Biomedical Physics (AIFB)·2026
Same author

Deep-stromal implantation of a double-crosslinked bioengineered porcine collagen implant (BPCDX) in advanced keratoconus: a 12-month study of corneal remodeling and visual outcomes.

Scientific reports·2026
Same author

Contributing Factors for Economic Inequality in Refractive Errors: An Analysis Employing Concentration Index and Oaxaca-Blinder Decomposition Methods in the Tehran Geriatric Eye Study.

Ophthalmic epidemiology·2026
Same author

Evolution of AI-based software as a medical device: from static tools to adaptive clinical partners.

Expert review of medical devices·2026

Related Experiment Video

Updated: Apr 23, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.0K

Predicting Keratoconus Progression From a Single Visit: Is Machine Learning Successful?

Alireza Jamali1, Hassan Hashemi, Morad Amir Ahmad

  • 1Rehabilitation Research Center (A.J., P.N.), Department of Optometry, School of Rehabilitation Sciences, Iran University of Medical Sciences, Tehran, Iran; Noor Ophthalmology Research Center (H.H.), Noor Eye Hospital, Tehran, Iran; Department of Optometry (M.A.A.), Department of physiotherapy, Erbil Technical & Medical Health Collage, Erbil Polytechnic University, Erbil, Kurdistan Region, Iraq; Department of Medical Radiation Engineering (F.B.M.), Science and Research Branch, Islamic Azad University, Tehran, Iran; Noor Research Center for Ophthalmic Epidemiology (A.H.), Noor Eye Hospital, Tehran, Iran; and Department of Medical Surgical Nursing (M.K.), School of Nursing and Midwifery, Shahid Beheshti University of Medical Sciences, Tehran, Iran.

Eye & Contact Lens
|April 21, 2026
PubMed
Summary
This summary is machine-generated.

Machine learning models can predict keratoconus (KCN) progression with high accuracy using longitudinal data. However, predicting KCN progression from a single visit remains challenging, requiring further validation for clinical use.

Keywords:
Artificial intelligenceClinical risk scoreCorneal tomographyKeratoconusMachine learningProgressionXGBoost

More Related Videos

Comparison of Agreement and Accuracy using Binocular Wavefront Optometer with Autorefractor and Phoropter
05:14

Comparison of Agreement and Accuracy using Binocular Wavefront Optometer with Autorefractor and Phoropter

Published on: September 16, 2025

812
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

2.2K

Related Experiment Videos

Last Updated: Apr 23, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.0K
Comparison of Agreement and Accuracy using Binocular Wavefront Optometer with Autorefractor and Phoropter
05:14

Comparison of Agreement and Accuracy using Binocular Wavefront Optometer with Autorefractor and Phoropter

Published on: September 16, 2025

812
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

2.2K

Area of Science:

  • Ophthalmology
  • Medical Artificial Intelligence
  • Corneal Diseases

Background:

  • Keratoconus (KCN) is a progressive corneal ectasia.
  • Accurate prediction of KCN progression is crucial for timely intervention.
  • Machine learning (ML) offers potential for predictive modeling in KCN.

Purpose of the Study:

  • To develop and evaluate ML models for predicting KCN progression in an Iranian cohort.
  • To assess the performance of models using longitudinal versus single-visit data.
  • To identify key features for KCN progression prediction.

Main Methods:

  • Retrospective study of 1,000 eyes from 529 KCN patients.
  • Progression defined by changes in KmaxF, astigmatism, pachymetry, or D-index.
  • Three XGBoost models were developed: longitudinal, single-visit, and refined binary.

Main Results:

  • The longitudinal ML model achieved near-perfect prediction (AUC=0.999).
  • Single-visit models showed limited accuracy (65%) and sensitivity (46%).
  • An optimized binary model improved sensitivity to 69.1% (AUC=0.72); key predictors included anterior curvature and thinnest pachymetry.

Conclusions:

  • ML models are highly effective with longitudinal KCN data.
  • Predicting KCN progression from single-visit data remains challenging.
  • Further validation and population-specific thresholds are needed for clinical implementation.