The Predictive Modeling for Early Pre-Eclampsia Identification

Abstract

Preeclampsia is a complex pregnancy complication characterized by high blood pressure and signs of damage to another organ system, most often the liver and kidneys. It typically occurs after 20 weeks of pregnancy and can lead to serious, even fatal, complications for both mother and baby if left unmanaged. Early prediction and intervention are crucial to managing the risks associated with preeclampsia. This Project explores the development and application of machine learning (ML) models to predict the likelihood of preeclampsia in pregnant women. Utilizing a dataset comprised of medical records, including demographic information, medical histories, and laboratory test results, we trained and evaluated several ML algorithms to identify those at high risk for developing preeclampsia. The project compares the performance of various models, including logistic regression, support vector machines, random forests, in terms of accuracy, sensitivity, and specificity. The best-performing model offers a promising tool for healthcare providers to enhance antenatal care by identifying high-risk patients early in their pregnancy, thereby enabling timely and targeted interventions. This research not only contributes to the field of medical informatics by advancing the predictive capabilities of ML in antenatal care but also demonstrates the potential for ML to improve outcomes in preeclampsia and other pregnancy-related complications.