Keywords
How to Cite
Abstract
Background: Maternal mortality and pregnancy-related complications continue to pose significant public health challenges in Pakistan, largely due to delayed diagnosis, inadequate continuous monitoring, and unequal access to quality antenatal healthcare. Recent advances in Artificial Intelligence (AI) have created opportunities for developing intelligent maternal health monitoring systems capable of predicting pregnancy complications through machine learning, predictive analytics, and clinical decision-support technologies. However, empirical evidence regarding the effectiveness of AI-enabled maternal health monitoring in the Pakistani healthcare context remains limited.
Objective: This study examined the effect of Artificial Intelligence-enabled maternal health monitoring on the prediction of pregnancy complications and maternal healthcare outcomes among pregnant women in Pakistan.
Methods: A quantitative cross-sectional research design was employed involving 450 pregnant women receiving antenatal care from selected public and private healthcare institutions in Pakistan. Data were collected using structured questionnaires, electronic medical records, and clinical assessments. Artificial Intelligence-based predictive models were developed using supervised machine learning algorithms to identify major pregnancy complications, including preeclampsia, gestational diabetes mellitus, hypertensive disorders, preterm birth, anemia, and fetal growth restriction. The proposed conceptual model was analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM), while the predictive performance of the AI model was evaluated using accuracy, precision, recall, F1-score, specificity, sensitivity, and Area Under the Receiver Operating Characteristic Curve (AUC-ROC).
Results: The findings revealed that Artificial Intelligence-enabled maternal health monitoring significantly improved the prediction of pregnancy complications and maternal healthcare outcomes (p < .001). The AI predictive model achieved an overall accuracy of 92.6% with an AUC-ROC of 0.95, demonstrating excellent predictive capability. Structural Equation Modeling further confirmed that AI-enabled monitoring significantly enhanced early identification of high-risk pregnancies, supported timely clinical intervention, improved antenatal care quality, and positively influenced maternal healthcare outcomes. Prediction of pregnancy complications also significantly mediated the relationship between AI-enabled maternal health monitoring and maternal healthcare outcomes.
Conclusion: Artificial Intelligence-enabled maternal health monitoring represents an effective digital healthcare strategy for strengthening maternal healthcare services in Pakistan. Integrating AI-driven predictive analytics, electronic health records, wearable technologies, and intelligent clinical decision-support systems into routine antenatal care can facilitate early diagnosis, optimize clinical decision-making, reduce pregnancy-related complications, and improve maternal and neonatal health outcomes. The findings provide important evidence for healthcare professionals, policymakers, and technology developers seeking to advance AI-enabled maternal healthcare in low- and middle-income countries.