Artificial Intelligence Sparse-Identification of Antenatal Corticosteroid Timing Effects Under Variable Ultrasound Gestational-Age Dating Error: A Causal Discovery Evaluation of Healthcare Neonatal Respiratory Safety

Ultrasound gestational-age dating error is a persistent obstacle to estimating the effect of antenatal corticosteroid timing on neonatal respiratory outcomes, because misclassification of exposure and outcome can bias both conventional regression and machine-learning estimators. The framework combines a measurement-error model for gestational age with a sparsity-constrained score-based causal discovery algorithm, and it is assessed using a before-after-control-impact design with a between-cluster standard deviation of 6.0, a residual standard deviation of 4.5, and an intraclass correlation of 0.64. Power was estimated across 2000 replications per design cell for cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points. The analysis indicates that 54 clusters are required to detect a 5-percentage-point effect at 80% power, and that the minimum detectable effect declines as cluster count increases. The framework is positioned as a methodological contribution to causal discovery in healthcare settings where gestational-age dating is imperfect, and its implications for neonatal respiratory safety evaluation are discussed.