Machine Learning-Based Approach to Predict Intrauterine Growth Restriction

Introduction: Creating a prediction model incorporating multiple risk factors for intrauterine growth restriction is vital. The current study employed machine learning to predict restriction. Methods: This cross-sectional was carried out in tertiary hospital Bandar Abbas, Iran, from January 2020 2022. Women with singleton pregnancies above the gestational age of 24 weeks who gave birth during period were included. Exclusion criteria included and fetal anomalies. Four statistical algorithms used build predictive model: (1) Decision Tree Classification, (2) Random Forest (3) Deep Learning, (4) Gradient Boost Algorithm. candidate predictors all models chosen based on expert opinion prior observational cohorts. To investigate performance each algorithm, some parameters, including area under receiver operating characteristic curve (AUROC), accuracy, precision, sensitivity, assessed. Results: Of 8683 women period, 712 recorded as having restriction, frequency 8.19%. Comparing parameters different showed that among four models, Learning had greatest an AUROC 0.91 (95% confidence interval, 0.85-0.97). importance variables revealed drug addiction, previous history chronic hypertension, preeclampsia, maternal anemia, COVID-19 weighted predicting Conclusions: A can be accurate algorithm