Development of a Model for Predicting Hypertension Disorders in Pregnancy Using Machine Learning

Hypertensive disorders during pregnancy (HDP) continue to be a significant contributor to maternal and foetal morbidity and mortality worldwide. Quick identification and accurate risk assessment are essential for reducing these health risks. This study centres on the development and implementation of a machine learning (ML) model designed to predict HDP risk, employing ML.NET and ASP.NET C#, in conjunction with a CSV-formatted clinical dataset. The model was trained with clinical factors like age, blood pressure, BMI, proteinuria, and diabetes status. We chose the Fast tree binary classification algorithm because it works well and is fast at classifying things. The model that was made was serialised and added to a web-based Patient Predictive Health Record System that was made with ASP.NET C# and linked to Microsoft SQL Server to handle patient information and user interactions. Using a DevOps approach, the system was built so that integration, testing, and deployment could happen all the time. We used standard machine learning metrics like accuracy, precision, recall, and AUC-ROC to test the model's performance through class-based analysis and 5-fold cross-validation. The results showed that the model was very good at making predictions, with an average cross-validation accuracy of 83.5%, an F1 score of 84.4%, and an AUC of 89.7%. In some folds, the accuracy reached 92.9%. These results confirm that the system is effective for clinical decision support in identifying HDP risk. This study promotes data-driven maternal healthcare by providing a scalable and practical solution for predicting HDP, which is well-suited for use in hospitals, antenatal clinics, and telemedicine services to help healthcare professionals provide timely and focused interventions.