Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jul 12, 2026

A Gaze-Contingent Display Framework for Perceptual Learning Research with Simulated Central Vision Loss

A Gaze-Contingent Display Framework for Perceptual Learning Research with Simulated Central Vision Loss

Published on: April 11, 2025

EGS-Net: a knowledge-augmented machine learning framework for predicting future high-myopia risk from longitudinal

Zhan Tang¹, Na Zhao², Zhaoyu Huang²

¹School of Physics, Zhejiang University, Hangzhou, China.

Frontiers in Medicine

|July 10, 2026

Summary

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

ATM counteracts chromatin-bound cGAS during DNA replication.

Nature cell biology·2026

Same author

Zearalenone causes female reproductive lipotoxicity through the ERα-CD36/TLR4 signaling pathway.

Communications biology·2026

Same author

Comprehensive CRISPR/Cas9-based mutagenesis identifies single-amino acid substitutions that abrogate SPEN function in X inactivation.

Nature communications·2026

Same author

Safety and feasibility of a novel bronchoplasty-based approach for the resection of middle mediastinal tumors located beneath the carina: a real-world study.

Journal of thoracic disease·2026

Same author

In situ multi-modal characterization of pancreatic cancer reveals tumor cell identity as a defining factor of the surrounding microenvironment.

Cell reports·2026

Same author

CeLR: A Transformer-Based Regression Network for Accurate Cephalometric Landmark Detection in High-Resolution X-Ray Imaging.

IEEE transactions on medical imaging·2026

A new Expert-Guided Stacking (EGS) model accurately predicts childhood myopia risk using longitudinal data. This AI approach enhances early detection and supports timely vision care interventions for students.

Area of Science:

Ophthalmology
Artificial Intelligence
Public Health

Background:

Childhood myopia is increasing globally, necessitating effective early risk detection tools.
Traditional screening methods using static data lack the ability to predict future refractive error progression.
Longitudinal refractive data is crucial for understanding dynamic changes and predicting high-myopia risk.

Purpose of the Study:

To develop and validate an Expert-Guided Stacking (EGS) predictive framework for childhood myopia risk stratification.
To leverage longitudinal school screening data for improved prediction of future high-myopia.
To create a scalable, knowledge-augmented machine learning solution for myopia surveillance.

Main Methods:

Utilized longitudinal school screening data from Binchuan County, China (Autumn 2023-Autumn 2025).

Keywords:

childhood myopia explainable machine learning future high-myopia risk prediction knowledge-augmented AI longitudinal school screening risk stratification

Related Experiment Videos

Last Updated: Jul 12, 2026

A Gaze-Contingent Display Framework for Perceptual Learning Research with Simulated Central Vision Loss

A Gaze-Contingent Display Framework for Perceptual Learning Research with Simulated Central Vision Loss

Published on: April 11, 2025

Developed an EGS framework combining a multi-model ensemble with a clinical risk-heuristic override.

Employed student-level partitioning, 5-fold cross-validation, and held-out test-set evaluation; SHAP analysis for interpretability.

Main Results:

The EGS framework achieved high clinical utility with Recall (0.9533) and Precision (0.9211).
Effectively identified future high-myopia risk while minimizing false positives compared to simpler models.
SHAP analysis confirmed the importance of longitudinal trajectory features for model interpretability.

Conclusions:

The developed EGS framework offers a robust and scalable solution for school-based myopia surveillance.
This AI-driven approach prioritizes high-risk individuals for timely clinical intervention and personalized vision care.
Knowledge-augmented machine learning can significantly enhance the accuracy and efficiency of myopia risk prediction.