Federated Semi-Supervised Graph Neural Networks with Prototype-Guided Pseudo-Labeling for Privacy-Preserving Gestational Diabetes Mellitus Prediction

TL;DR

This paper introduces FedTGNN-SS, a federated semi-supervised graph neural network framework that leverages prototype-guided pseudo-labeling and privacy-preserving techniques for early prediction of gestational diabetes using electronic health records.

Contribution

It proposes a novel federated semi-supervised GNN approach with prototype-guided pseudo-labeling and privacy-safe prototype sharing for GDM prediction.

Findings

Achieves 56 significant wins over federated baselines.

Attains high AUROC scores under extreme label scarcity.

Effective across multiple diabetes-related datasets.

Abstract

Gestational Diabetes Mellitus (GDM) is a high-prevalence pregnancy complication that requires accurate early risk stratification to reduce maternal and fetal morbidity. However, real-world clinical deployment of machine learning is hindered by two coupled constraints: (i) label scarcity, where a large fraction of electronic health records (EHR) lack confirmed diagnostic labels, and (ii) data privacy, which prevents sharing patient-level data across hospitals. This paper proposes FedTGNN-SS, a privacy-preserving federated semi-supervised framework for clinical tabular EHR. Each hospital builds a local k-nearest-neighbor patient similarity graph and trains a topology-adaptive GNN encoder. To robustly exploit unlabeled records, FedTGNN-SS combines (1) prototype-guided pseudo-labeling with neighborhood agreement, (2) adaptive graph refinement that periodically updates the k-NN graph using learned embeddings, (3) clinical-aware consistency augmentation applied only to continuous variables, and (4) privacy-safe prototype sharing that exchanges only class-level centroids. Across three diabetes-related datasets (GDM: N = 3,525; Pima: N = 768; Early Stage: N = 520) under 10\%-80\% missing labels per silo, FedTGNN-SS achieves 56 significant wins ($p < 0.05$) against 11 federated baselines and attains strong AUROC under extreme scarcity (Pima: 0.8037 at 80\% missing, Early Stage: 0.9634 at 80\% missing).