Abstract
This study developed and evaluated an Artificial Intelligence (AI)-assisted predictive framework for identifying high-risk pregnancy outcomes in rural and urban healthcare settings of Pakistan. Despite improvements in maternal healthcare services, Pakistan continues to experience high rates of maternal and neonatal complications due to delayed risk identification, limited healthcare resources, and disparities between rural and urban healthcare infrastructure. To address this issue, a quantitative predictive analytics design was employed using retrospective maternal healthcare data obtained from multiple healthcare institutions. The dataset included demographic, clinical, socioeconomic, and healthcare accessibility variables, which were analyzed using multiple machine learning algorithms including Random Forest, XGBoost, Support Vector Machine (SVM), Artificial Neural Network (ANN), and Long Short-Term Memory (LSTM) models. The results indicated that AI-based models achieved high predictive accuracy in identifying high-risk pregnancies, with LSTM and XGBoost outperforming other algorithms. Clinical indicators such as blood pressure, blood glucose levels, BMI, and previous pregnancy complications emerged as the most significant predictors of adverse pregnancy outcomes. Comparative analysis further revealed slightly higher predictive performance in urban healthcare settings compared to rural settings; however, overall model performance remained strong across both contexts. The findings confirmed that AI-assisted predictive systems significantly enhance early risk detection and support clinical decision-making in maternal healthcare.
The study concludes that integrating AI-driven clinical decision support systems into maternal healthcare services can improve early identification of high-risk pregnancies, optimize resource allocation, and contribute to improved maternal and neonatal outcomes in Pakistan. The findings provide valuable implications for healthcare practitioners, policymakers, and researchers aiming to strengthen digital health transformation in maternal care