This study aims to perform statistical inference and prediction of changes in fetal heart rate during active labor in healthy pregnant women by comparing three different machine learning methods

Purpose: This study aims to perform statistical inference and prediction of changes in fetal heart rate during active labor in healthy pregnant women by comparing three different machine learning methods. Methods: A retrospective analysis of 1077 healthy laboring parturients receiving neuraxial analgesia was conducted. We compared a principal components regression model with treebased random forest, ridge regression, multiple regression, a general additive model, and elastic net in terms of prediction accuracy and interpretability for inference purposes.

Inclusion Criteria:

• Older than 18 years
• Pregnancy requiring labor analgesia
• Active labor
• Request of neuraxial analgesia per patient and/or obstetrician
• Received combined spinal-epidural technique

Exclusion Criteria:

• Uterine tachysystole before neuraxial analgesia.
• Baseline blood pressure <90/60 mmHg.
• Third trimester hemorrhage
• Eclampsia
• Allergies to local anesthetics or fentanyl.
• Maternal fever.
• Pruritus before performance of neuraxial analgesia

-Pregnant patients admitted to the labor and delivery unit of an academic center