Medical Q&A Chatbot

A modern medical question-answering chatbot that provides appropiate, relevant, and user-friendly medical information using a combination of retrieval-based methods and large language models.

Features

• Hybrid Retrieval System: Combines MedQuAD dataset with Meditron LLM for accurate medical information
• Enhanced Responses: Uses Llama 3.1 (8B) to improve clarity and readability of medical information
• Entity Recognition: Highlights important medical terms and concepts
• Conversational Interface: Natural conversation capabilities for a better user experience
• Responsive Design: Clean, modern UI with Tailwind CSS
• Loading Indicators: Visual feedback during response generation

Architecture

The application consists of two main components:

Backend (FastAPI)

• Query Processing: Processes and optimizes user questions
• Entity Recognition: Identifies medical entities in text
• Retrieval Service: Searches the MedQuAD dataset for relevant information
• LLM Integration: Uses Meditron for medical knowledge and Llama 3.1 for response enhancement
• Conversation Service: Handles basic conversational interactions

Frontend (React + TypeScript)

• Chat Interface: Modern, responsive chat UI
• Message Formatting: Properly formats and displays medical information
• Entity Highlighting: Visually highlights medical terms
• Loading States: Provides visual feedback during processing

Project Structure

medical-qa-chatbot/
├── backend/               # FastAPI backend
│   ├── app/
│   │   ├── api/           # API endpoints
│   │   ├── services/      # Business logic
│   │   ├── utils/         # Utility functions
│   │   └── main.py        # Application entry point
│   └── run.py             # Server startup script
├── frontend/              # React frontend
│   ├── src/
│   │   ├── components/    # React components
│   │   ├── services/      # API services
│   │   ├── types/         # TypeScript types
│   │   └── App.tsx        # Main application component
│   └── index.html         # HTML entry point
└── data/                  # Data files and indexes

Setup and Installation

Prerequisites

• Node.js (v16+)
• Python (v3.9+)
• Ollama (for running local LLMs)

Backend Setup

1. Navigate to the backend directory:

   cd backend
   

2. Create a virtual environment:

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate
   

3. Install dependencies:

   pip install -r requirements.txt
   

4. Start the backend server:

   python run.py
   

Frontend Setup

1. Navigate to the frontend directory:

   cd frontend
   

2. Install dependencies:

   npm install
   

3. Start the development server:

   npm run dev
   

LLM Setup

1. Install Ollama from https://ollama.ai/

2. Pull the required models:

   ollama pull meditron:7b
   ollama pull llama3.1:8b
   

Usage

1. Open your browser and navigate to http://localhost:5173
2. Type a medical question in the input field
3. The system will:
   • Process your question
   • Search for relevant medical information
   • Enhance the response for better readability
   • Display the answer with highlighted medical terms

Response Types

The chatbot provides different types of responses:

• Medical Information: Structured, concise answers to medical questions
• Conversational Responses: Natural replies to greetings and simple questions
• Not Available Responses: Clear indication when information is not available

Technologies Used

• Frontend: React, TypeScript, Tailwind CSS, Vite
• Backend: FastAPI, Python
• LLMs: Meditron 7B, Llama 3.1 8B (via Ollama)
• Data: MedQuAD (Medical Question Answering Dataset)

Disclaimer

This chatbot provides information based on the MedQuAD dataset and AI language models. It is not a substitute for professional medical advice, diagnosis, or treatment. Always consult with a qualified healthcare provider for medical concerns.