RAG-Powered Healthcare Assistant is an AI-driven tool using Retrieval-Augmented Generation (RAG) with multilingual support (Darija, Arabic, English) to provide personalized health info. It features Prescription OCR, multilingual queries, and a Pharmacy Locator to offer tailored, accurate healthcare responses and services.
# RAG-Powered Healthcare Assistant
An AI-driven healthcare assistant that utilizes Retrieval-Augmented Generation (RAG) with multilingual support in Darija, Arabic, and English to offer personalized health management. The app helps users with prescription scanning, health queries, multilingual support, and pharmacy locator.
- Watch the video demo of our app video-demo.
## Background and Problem Statement
Managing healthcare information effectively remains a challenge in today's world. Common issues include:
- Fragmented medical records
- Difficulty in obtaining personalized medical advice
- Language barriers in healthcare communication
- Difficulty locating nearby healthcare providers
## Impact and Proposed Solution
The **RAG-Powered Healthcare Assistant** aims to solve these challenges by:
- Centralizing personal health records for easy access
- Leveraging RAG to provide intelligent, context-aware responses to health queries
- Enabling Prescription OCR for easy scanning and processing of prescriptions
- Offering multilingual support (Darija, Arabic, English) for diverse users
- Incorporating a Pharmacy Locator with geolocation to help users find nearby pharmacies and healthcare providers
## Project Outcomes and Deliverables
This project delivers:
- A fully functional AI-powered healthcare assistant
- A multilingual chatbot for health queries
- An OCR system for prescription scanning and processing
- An integrated Pharmacy Locator with real-time geolocation
- Secure storage and retrieval of personalized medical records
## Technology Stack
- **Frontend**:
- Streamlit - Main web interface
- Chainlit - Chat interface
- **Backend**:
- Python 3.9+
- LangChain - For AI/LLM orchestration
- ChromaDB - Vector database for document storage
- Ollama - Local LLM integration
- OpenCV - Image processing
- PyPDF & PDFPlumber - PDF processing
- Unstructured - Document parsing
- ElevenLabs - Text-to-speech capabilities
## Prerequisites
Before running the application, …