Artificial Intelligence Conformal Prediction of Neonatal Jaundice Progression Under Variable Transcutaneous Bilirubin Measurement Interval Irregularity: A Distribution-Free Evaluation of Healthcare Readmission Risk

Neonatal jaundice remains a leading cause of hospital readmission, and transcutaneous bilirubin (TcB) monitoring is frequently irregular in community and outpatient settings, producing measurement intervals that vary unpredictably across infants. The framework treats each infant's TcB trajectory as a functional observation indexed by irregular time points, constructs nonconformity scores from a weighted functional regression, and calibrates prediction intervals via split conformal inference without distributional assumptions. A before-after-control-impact design with an intraclass correlation of 0.64, between-cluster standard deviation of 6.0 and residual standard deviation of 4.5 was used to characterise power across 8, 16, 24 and 32 clusters and effect sizes of 3, 5 and 8 percentage points, with 2000 replications per cell. Power reached conventional levels at 56 clusters for a 5-percentage-point effect. The framework provides valid marginal coverage under interval irregularity and offers a principled basis for readmission-risk stratification where measurement schedules cannot be standardised.