Leveraging digital health and machine learning models to forecast adverse maternal outcome in low resource settings, Geita Tanzania

Tanzania’s maternal mortality ratio remains alarmingly high at 238 deaths per 100,000 live births. To address this, PHIT, with support from the Bill & Melinda Gates Foundation, launched a two‑year project using digital health and machine learning to predict adverse maternal outcomes. A tablet‑based system streamlined ANC data collection, and over 187,000 client records were used to train an XGBoost model. The model achieved 99% accuracy and 43% precision in detecting gestational hypertension risk, showing the power of AI to ease provider workload and improve early detection in low‑resource settings.