Predictive Technology in Maternal Health: artificial Intelligence Models for the Identification of Obstetric Risks

Introduction: obstetric complications are one of the leading causes of maternal morbidity and mortality worldwide. Artificial intelligence (AI) has proven to be an effective tool for predicting obstetric risks, enabling timely interventions. However, its implementation in countries with limited healthcare infrastructure remains a challenge.Methods: a predictive AI-based model was developed using clinical data from 2 500 pregnant women treated in Ecuadorian hospitals. Logistic regression, neural networks, and random forest algorithms were evaluated to predict complications such as preeclampsia, preterm birth, and gestational diabetes. Cross-validation techniques and inferential statistical analyses were applied.Results: neural networks demonstrated the best performance, with an accuracy of 92 % and an AUC-ROC of 0.94, outperforming traditional models. The main risk factors identified were high blood pressure, high body mass index, and family history.Conclusions: aI can significantly improve the early detection of obstetric complications, especially in resource-limited settings. Implementing these models in hospital systems would help optimize maternal-fetal care and reduce maternal mortality.