Ndile317/maternal-health-cdss

AI-Powered Clinical Decision Support System for Maternal Health in Zimbabwe" # 🤰 Maternal Health Clinical Decision Support System An AI-powered clinical decision support system for predicting adverse maternal and neonatal outcomes in Zimbabwe. ## 🚀 Live Application **Deployed on Streamlit Cloud:** maternal-health-cdss.stream… ## 🏥 Clinical Features - **AI Risk Prediction**: Machine learning model for adverse outcome prediction - **SHAP AI Explanations**: Transparent AI with feature importance analysis - **Clinical Recommendations**: Evidence-based intervention guidance - **Patient Management**: Complete electronic health records system - **Analytics Dashboard**: Population-level insights and reporting ## 🔬 Research Basis - **Model**: Random Forest Classifier (AUC: 0.88) - **Training Data**: Zimbabwe Demographic and Health Survey 2015 - **Validation**: Multiple Indicator Cluster Survey 2019 - **Top Predictors**: ANC Quality Score, Maternal Age, Parity - **SHAP Integration**: Full explainable AI implementation ## 🛠️ Technology Stack - Python 3.11 - Streamlit (Web Framework) - Scikit-learn (Machine Learning) - SHAP (Explainable AI) - Plotly (Interactive Visualizations) - Pandas (Data Processing) ## 📊 Model Performance - **AUC**: 0.88 (Excellent discriminative ability) - **Top Feature**: ANC Quality Score (30.2% importance) - **Validation**: Cross-dataset validation performed - **SHAP**: Full model interpretability ## 📁 Project Structure \\\ maternal_health_cdss/ ├── app.py # Main Streamlit application ├── requirements.txt # Python dependencies ├── setup.sh # Streamlit deployment configuration ├── .gitignore # Git ignore rules ├── README.md # Project documentation └── utils/ # Utility modules ├── __init__.py ├── model_loader.py # AI model with SHAP integration └── risk_calculator.py # Clinical risk calculation \\\ ## 🚀 Local Development \\\ash # Install dependencies pip install -r requirements.txt # Run application locally streamlit run app …