Development and evaluation of machine learning models for individualized prediction of myopia control efficacy treated with overnight orthokeratology - Original Research

Q: What is the study’s primary objective?

To develop a predictive model using routine clinical and ocular measurements to forecast the effect of overnight orthokeratology on myopia control efficacy, helping ophthalmologists choose appropriate methods for adolescents.

Q: How was the dataset constructed and which approach was used to select predictive factors?

A retrospective cohort with one-year follow-up data from 225 orthokeratology-treated myopia children was randomly split into training (n=180) and test (n=45). LASSO regression identified predictive factors correlated with myopia control.

Q: Which machine learning models were compared, and what model showed the best predictive performance?

KNN, SVM, RF, Extra Trees, and XGBoost were compared using accuracy, ROC, and DCA curves. SVM achieved the highest predictive quality, with AUC values of 0.877 in training and 0.828 in external validation.

Aug 4, 2026

Upload by Rowan

Development and evaluation of machine learning models for individualized prediction of myopia control efficacy treated with overnight orthokeratology. Using retrospective one-year follow-up data from 225 adolescent myopia children, clinical and corneal topography variables were screened with LASSO regression to identify predictive factors. Multiple models were compared by accuracy, ROC/AUC, and decision curve analysis to estimate clinical net benefit across threshold ranges. The study builds practical tools to forecast 1-year axial length elongation and support timely adjustment of orthokeratology-based treatment while guiding lens optimization.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

Introduction

Myopia burden and risks

Orthokeratology and evidence

Mechanisms and research gaps

Methods

Data and participants

Feature selection and modeling

Evaluation metrics

Results

Selected predictive features

Model comparisons and validation

Clinical benefit via DCA

Conclusion

OPEN ACCESS EDITED BY Sheila Gillard Crewther, La Trobe University, Australia REVIEWED BY Amira Awad Moawad, Friedrich Loeffler Institut, Germany Brian Vohnsen, University College Dublin, Ireland *CORRESPONDENCE Qi Zhao [dalianzhaoqi@126.com](dalianzhaoqi@126.com) RECEIVED 12 January 2025 ACCEPTED 04 April 2025 PUBLISHED 12 May 2025 CITATION Zhang L, Gao M, Wang Y, Zhang S, Zhu H and Zhao Q (2025) Development and evaluation of machine learning models for individualized prediction of myopia control efficacy treated with overnight orthokeratology. Front. Med. 12:1559435 . doi: 10.3389/fmed.2025.1559435 COPYRIGHT © 2025 Zhang, Gao, Wang, Zhang, Zhu and Zhao. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. TYPE Original Research PUBLISHED 12 May 2025 DOI 10.3389/fmed.2025.1559435 Development and evaluation of machine learning models for individualized prediction of myopia control efficacy treated with overnight orthokeratology Lan Zhang, Mingjun Gao, Yiru Wang, Siqi Zhang, Huailin Zhu and Qi Zhao * Department of Ophthalmology, The Second Hospital of Dalian Medical University, Dalian, China Purpose: The primary objective of this study is to develop a predictive model utilizing fundamental clinical and ocular measurements to predict the effect of overnight orthokeratology on myopia control. Accordingly, this study aims to assist ophthalmologists in selecting adolescent myopia control methods. Methods: This retrospective study used one-year follow-up data of 225 myopia children treated with orthokeratology. Using the random sampling method, 225 samples were randomly divided into a training set (n = 180) and a test set (n = 45) . LASSO regression identified predictive factors correlated with controlling myopia. The final features are input into the machine learning model for prediction model construction to predict 1-year axial length elongation. The prediction performance was evaluated according to the accuracy and AUC of the training set and the test set. DCA was used to assess the clinical benefits of the model. Results: Five features (age, diopter, flat keratometry, corneal higher-order aberrations (6 mm), and intraocular trefoil (6 mm) were used to build the machine learning model (p < 0.01)) . Based on the accuracy, ROC, and DCA curves, the prediction performance and clinical practicability of five prediction models: KNN, SVM, RF, Extra Trees, and XGBoost were compared. In the DCA, all machine learning models consistently achieved greater net benefits within the clinical threshold range. SVM demonstrated the highest predictive quality with an AUC of 0.877 in the training and 0.828 in the external validation set. Conclusion: We developed and validated several prediction models for individualized prediction of myopia control efficacy treated with overnight orthokeratology through machine learning, using easily obtained clinical and corneal topography features. This cost effective strategy helps ophthalmologists predict the effect of using orthokeratology in children, and make timely adjustments to myopia control methods. The differential features selected by this model can also provide insights for optimizing lens design. KEYWORDS machine learning, myopia, orthokeratology, axial length, high-order aberration Frontiers in Medicine 01 [frontiersin.org](frontiersin.org) Introduction In recent years, the prevalence of myopia in adolescents has been on the rise, especially in urbanized areas of Asia. In these areas, the myopia prevalence in young people is about 80 -90%, and

FAQs

What is the study’s primary objective?

To develop a predictive model using routine clinical and ocular measurements to forecast the effect of overnight orthokeratology on myopia control efficacy, helping ophthalmologists choose appropriate methods for adolescents.

How was the dataset constructed and which approach was used to select predictive factors?

A retrospective cohort with one-year follow-up data from 225 orthokeratology-treated myopia children was randomly split into training (n=180) and test (n=45). LASSO regression identified predictive factors correlated with myopia control.

Which machine learning models were compared, and what model showed the best predictive performance?

KNN, SVM, RF, Extra Trees, and XGBoost were compared using accuracy, ROC, and DCA curves. SVM achieved the highest predictive quality, with AUC values of 0.877 in training and 0.828 in external validation.

Aug 4, 2026

Upload by Rowan