Albyaz/african-health-pipeline

# African Health Data Pipeline 🌍 A production-style ETL data pipeline that extracts real health data from the WHO Global Health Observatory API, transforms and validates it, and loads it into a cloud PostgreSQL database (Neon). Built by **Albright Ndamati** — Medical Doctor and Data Engineer --- ## Why I Built This As a medical graduate, I've seen firsthand how poor health data infrastructure affects healthcare delivery across Africa. This project combines my clinical background with data engineering skills to build a system that makes African health metrics accessible, clean, and queryable. This project is also foundational to **DAAL** — my preventive digital health platform for Africa. --- ## What It Does Extracts real WHO health data for 15 African countries including: - Nigeria, Ghana, Kenya, South Africa, Ethiopia - Tanzania, Uganda, Senegal, Zambia, Zimbabwe - Cameroon, Angola, Mozambique, Egypt, Ivory Coast Tracks 3 key health indicators: - Life Expectancy (2000-2024) - Child Mortality Rate (2000-2024) - Malaria Incidence (2000-2024) **Total: 4,245 validated records loaded into cloud PostgreSQL** --- ## Pipeline Architecture WHO GHO API → extract.py → transform.py → load.py → Neon PostgreSQL ↓ ↓ ↓ raw_data.csv clean_data.csv health_metrics table Run the full pipeline with one command: ```bash python pipeline.py ``` --- ## Tech Stack - **Python 3.11** — Core language - **Pandas** — Data transformation and validation - **SQLAlchemy** — Database ORM - **PostgreSQL (Neon)** — Cloud database - **WHO GHO API** — Real health data source - **Git** — Version control --- ## Project Structure african_health_pipeline/ ├── extract.py # Pulls data from WHO API ├── transform.py # Cleans and validates data ├── load.py # Loads into PostgreSQL ├── pipeline.py # Runs full ETL pipeline ├── .env # Database credentials (not committed) ├── .gitignore └── README.md --- ## How to Run ```bash # Clon …