Risk Assessment Using a Machine Learning-Based Mobile Application for Gestational Hypertension in Lima, Peru

Ramirez J.R.; Arroyo J.H.; Cerna E.A. · Proceedings of the 2025 IEEE 5th International Conference on Advanced Learning Technologies on Education and Research, ICALTER 2025 (2025) · DOI: 10.1109/icalter69698.2025.11355100

Resumen

Gestational hypertension is a leading cause of maternal morbidity and mortality, affecting about 10% of pregnant women. Limitations in access to prenatal care increase the likelihood of late diagnosis. This work proposes a mobile application that integrates machine learning models and business intelligence techniques to support early detection of hypertensive risk during pregnancy. The system enables pregnant women to record vital signs, receive preventive alerts, and view intuitive comparisons that reinforce their active role in health monitoring. The Random Forest-based model used for risk level calculation achieved an accuracy of 0.923. Validation showed a more than 20% increase in pregnant women's willingness to seek immediate clinical follow-up after a high-risk alert compared with 2024 reports. These findings demonstrate the reliability and applicability of the solution, with potential to complement in-person care and strengthen maternal self-care in global contexts.