FNNtechdev/Afyaguide

AI-powered health facility recommender for Kenya # 🏥 AfyaGuide ### **From “Where should I go?” to a clearer, data-informed healthcare choice. 🇰🇪** > **AfyaGuide is a healthcare-facility discovery and recommendation prototype for Kenya.** > > It lets users describe a healthcare need in everyday language, identifies the most relevant standardized service, and ranks potentially suitable facilities using **service availability, facility capability, and geographic proximity**. --- ## ✨ Why AfyaGuide? Finding the *right* health facility can be harder than finding a facility. A user may know: > *“I need HIV testing near me.”* but not know the standardized service name, which facility level can provide it, or which suitable facility is closest. **AfyaGuide bridges that gap.** ### The idea in one line ```text Everyday language → Health-service understanding → Suitable facilities → Location-aware recommendations ``` --- ## 🎯 The Problem Healthcare information is often organized around **facility names, service labels, and administrative terminology** rather than the way people naturally describe their needs. This creates a practical discovery problem: - ❓ Which service do I actually need to look for? - 🏥 Which facilities provide it? - 📊 Does the facility have the required capability? - 📍 Which suitable options are near me? - 🧭 How can I compare the available options? AfyaGuide was developed to make this first step of healthcare navigation simpler. > **Important:** AfyaGuide is a facility-discovery and decision-support tool. It is **not a diagnostic system** and does not replace qualified medical advice. --- # 🌍 What AfyaGuide Does | Capability | What it does | |---|---| | 🗣️ **Natural-language search** | Lets users describe their healthcare need in ordinary language | | 🧠 **Health-need matching** | Maps the request to a standardized health need/service | | 🏥 **Service matching** | Identifies facilities that appear to offer the relevant service | | 📊 **KEPH filtering** | Considers the minimum required fa …