EGS-Net: a knowledge-augmented machine learning framework for predicting future high-myopia risk from longitudinal school-screening trajectories - Takeaways - MDSpire
Coming Soon: Introducing MDSpire News. Learn more
Conexiant’s news site is now MDSpire News. Learn more

EGS-Net: An Enhanced Machine Learning Approach for Forecasting High-Myopia Risk Based on Longitudinal School Screening Data

  • By

  • Zhan Tang

  • Na Zhao

  • Zhaoyu Huang

  • Jinhao Lu

  • Chao Dai

  • Jian Wang

  • Runze Zheng

  • June 25, 2026

Share

  • 1

    The EGS predictive framework utilizes longitudinal school screening data to enhance early risk stratification for high myopia.

  • 2

    The model integrates multi-model ensemble learning with a clinical risk-heuristic override to improve prediction accuracy.

  • 3

    EGS-Net achieved high Recall (0.9533) and Precision (0.9211), enabling reliable identification of future high-myopia risk.

  • 4

    SHAP analysis confirmed the importance of longitudinal trajectory features for model interpretability and transparency.

  • 5

    The study addresses gaps in existing models by leveraging individualized kinetic signals and integrating clinical domain knowledge.

Original Source(s)

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

  • by Zhan Tang, Na Zhao, Zhaoyu Huang, Jinhao Lu, Chao Dai, Jian Wang, Runze Zheng

  • June 25, 2026

Related Content

Modeling Oxygen Saturation Targets, Mortality, and Retinopathy Among Extremely Preterm Infants

Pinhole Optics Meet Cataract Surgery

Small-aperture IOL shows promise for cataract patients with pre-existing corneal comorbidities

  • July 28, 2026

  • 3 min
GA Treatment Preferences

Survey questions what drives geographic atrophy treatment decisions for clinicians

  • August 4, 2026

  • 4 min