The Role of Artificial Intelligence in Predicting Intrauterine Fetal Demise: A Case-Control Study

Abstract

Intrauterine Fetal Demise (IUFD) remains a major cause of perinatal mortality, often occurring unexpectedly despite routine antenatal surveillance. Traditional clinical and imaging parameters are limited in their predictive precision, underscoring the need for advanced analytical frameworks. This study investigates the role of Artificial Intelligence (AI) in predicting IUFD through a case-control design involving 500 pregnancies (250 IUFD cases and 250 matched controls). Clinical, biochemical, and ultrasonographic parameters such as maternal age, body mass index, blood pressure, fetal heart rate variability, Doppler indices, and placental characteristics were analyzed. Machine learning models including Logistic Regression, Random Forest, and Gradient Boosting were trained to classify IUFD risk using retrospective hospital data. The Random Forest model achieved the highest performance with an accuracy of 92%, sensitivity of 90%, and an area under the curve (AUC) of 0.94. Key predictors identified included abnormal Doppler flow ratios, reduced fetal movement scores, and elevated maternal diastolic pressure. These findings demonstrate that AI-based predictive modeling can provide significant clinical insights, allowing for early risk stratification and targeted interventions. Integrating AI into obstetric care could transform fetal monitoring from reactive diagnosis to proactive prevention, ultimately reducing fetal mortality in high-risk pregnancies.