AI Algorithms Predict Gestational Diabetes Accurately

A systematic review and meta-analysis by Liang et al., published in J Med Internet Res (2026), evaluated AI-based prediction models for gestational diabetes using studies through June 1, 2025, identifying 14 eligible studies. Pooled sensitivity was 0.78 and specificity 0.85 with AUC 0.94; random forest showed highest sensitivity (0.83) and XGBoost highest specificity (0.82), but small-study effects and heterogeneity warrant local validation and larger prospective studies.