Explainable Health Risk Prediction Using Lifestyle and Wearable Data with Fine-Tuning and RAG

An explainable health risk prediction system that uses lifestyle and wearable data with fine-tuning and RAG.

An explainable health risk prediction system that uses lifestyle and wearable data with fine-tuning and RAG. As a reference project it shows how a system like this is put together in practice, which parts matter and how they are tested, so that a developer can learn from it and build on it.

It suits learners who want to connect hardware to software: reading sensors, sending data to a back end, visualising it and acting on it with alerts or control.

The main parts are explanations of each result (feature importance or visual evidence), retrieval over an indexed document collection that grounds the language model's answers, sensor or device data collection, a messaging layer between devices and the server, and a live dashboard. The reference build uses LangChain, FAISS, SHAP, ESP32, and MQTT, with a clear project structure, setup instructions and sample data so that it can be run and extended.