SafeMom is an AI-driven clinical dashboard designed to reduce maternal mortality in low resource settings.
SafeMom: AI-Powered Maternal Triage
SafeMom is a clinical decision-support tool designed to reduce maternal mortality in low-resource settings. It uses machine learning to stratify pregnancy risks based on real-time clinical vitals.
- The Mission
In regions with low midwife-to-patient ratios, delayed triage is a leading cause of maternal death. SafeMom provides an instant risk score, allowing healthcare workers to prioritize high-risk patients for immediate referral.
- Technical Stack
Engine: LightGBM (Gradient Boosting) calibrated for clinical sensitivity.
Frontend: Streamlit-based "High-Agency" Dashboard.
Intelligence: Integrated Risk Driver Analysis (identifying Sepsis, Preeclampsia, and Anemia flags).
Project Structure
src/: Production-ready Streamlit application.
notebooks/: Model training, data cleaning, and validation benchmarks.
assets/: Saved model weights (.pkl) and UI branding.
🛑 Disclaimer
This is a Baseline MVP for research and demonstration purposes. It is intended to assist, not replace, professional clinical judgment.
## 👥 Collaborators
* **Becky (
github.com)**