Research Article: SSR-stacking: a hybrid GNN–ensemble framework for high-myopia classification under simulated baseline-SE missingness
Abstract:
The increasing burden of childhood myopia creates a need for screening models that remain informative when routinely collected clinical measurements are incomplete. We propose SSR-Stacking, a hybrid framework integrating Graph Neural Networks (GNN) with Ensemble Learning for classification of final-follow-up high-myopia status under simulated missingness of baseline spherical equivalent (SE). Unlike traditional models, we construct a Social-Environment Graph based on school-class affiliations to represent classroom-level relational structure. Our framework features a Self-Supervised Reconstruction (SSR) mechanism toreconstruct randomly masked baseline SE from neighboring peers and a Meta-Stacking layer to fuse GNN embeddings with diverse ML classifiers (XGBoost, SVM, RF). Evaluated on a longitudinal dataset ( N = 4, 973) from Binchuan, China, under 50% simulated baseline-SE missingness using classroom-level grouped folds and out-of-fold stacking, SSR-Stacking achieved a ROC-AUC of 0.9184 (95% CI: 0.9062–0.9306), PR-AUC of 0.6120, Recall of 0.8119, Specificity of 0.9851, and F1-score of 0.7506 on unsampled held-out predictions with a natural high-myopia prevalence of 4.06%. These internal results support further evaluation under natural missingness and external validation before clinical deployment.
Introduction:
Large-scale school-based ocular screening is an important component of childhood myopia surveillance ( 1 – 4 ). However, predictive models developed from complete datasets may be difficult to apply when routine measurements are missing. In regional screening programs, biometric and refractive records may be incomplete because of measurement failure, student absence, or heterogeneous administrative systems. Consequently, models that assume a fully observed feature matrix may experience substantial performance…
Read more