README.md

---
title: CardioQA AI System
emoji: 🫀
colorFrom: blue
colorTo: red
sdk: docker
pinned: false
license: mit
---

# CardioQA - AI-Powered Cardiac Diagnostic Assistant 🫀

Professional cardiac health assistant powered by **RAG** (Retrieval-Augmented Generation), **ChromaDB**, and **Google Gemini 2.0**.

## 🌟 Features
- 🫀 **364+ cardiac medical Q&As** - Comprehensive knowledge base
- 🤖 **AI-powered responses** with Google Gemini 2.0 Flash
- 🛡️ **Medical safety validation** - Emergency detection & warnings  
- 📊 **Confidence scoring** - Dynamic response quality metrics
- ⚡ **Real-time semantic search** - ChromaDB + Sentence Transformers
- 🎯 **Professional safety features** - Medical disclaimers & consultation recommendations

## 🚀 API Endpoints
- **`/`** - API information & status
- **`/health`** - System health check & database status
- **`/query`** - Ask cardiac health questions (POST)
- **`/docs`** - Interactive FastAPI documentation
- **`/stats`** - System statistics & model information

## 🛠️ Built With
- **Backend:** FastAPI + Uvicorn
- **AI Model:** Google Gemini 2.0 Flash
- **Vector Database:** ChromaDB with persistent storage
- **Embeddings:** Sentence Transformers (all-MiniLM-L6-v2)
- **Safety:** Custom medical validation system
- **Deployment:** Docker + HuggingFace Spaces

## 📊 System Architecture
User Query → Semantic Search (ChromaDB) → Context Retrieval →
Gemini AI Generation → Safety Validation → Response + Disclaimers


## ⚠️ Medical Disclaimer
This system provides **educational information only**. Always consult qualified healthcare professionals for medical advice, diagnosis, or treatment. For medical emergencies, contact emergency services immediately.

## 👨‍💻 Developer
Built by **Novonil Basak** - AI/ML Engineer & Biotechnology Student

---