🩺 MedBot - AI Medical Chatbot using RAG

📖 About the Project

MedBot is an AI medical chatbot that answers health-related questions using information from a medical book.

Instead of answering only from the AI model's general knowledge, the chatbot first searches the medical book for the most relevant information. It then uses that information to generate an answer. This approach is called Retrieval-Augmented Generation (RAG).

I built this project to learn how Large Language Models (LLMs), LangChain, Vector Databases, and Prompt Engineering work together in a real application.

🚀 Live Demo

🌐 Try MedBot: medbot

📂 GitHub Repository: https://github.com/OmkarSahu0375/medbot

📷 Screenshot

Home Screen

Chat Example

✨ Features

Answer medical questions using a medical knowledge base
Retrieve relevant information before generating an answer (RAG)
Semantic search using Pinecone Vector Database
Stream responses in real time
Remember previous messages during the conversation
Simple and clean Streamlit interface
Faster loading using Streamlit caching

🛠️ Tech Stack

Category	Technology
Language	Python
Frontend	Streamlit
LLM	GPT-OSS-20B (Groq)
AI Framework	LangChain
Embedding Model	sentence-transformers/all-MiniLM-L6-v2
Vector Database	Pinecone
Document Loader	PyPDF

🏗️ System Architecture

The chatbot follows these steps to answer a user's question:

Medical PDF files are loaded and split into smaller chunks.
Each chunk is converted into embeddings using the MiniLM embedding model.
The embeddings are stored in Pinecone Vector Database.
When a user asks a question, the question is converted into an embedding.
Pinecone searches for the most relevant medical information.
The retrieved information, along with the user's question and conversation history, is sent to the LLM.
The LLM generates the final answer and streams it back to the user in real time.

📁 Project Structure

medbot/
│
├── data/
│   └── medical-book.pdf
│
├── src/
│   ├── helper.py
│   └── prompt.py
│
├── streamlit_app.py
├── requirements.txt
├── README.md
└── .env

🚀 Installation

Clone the repository

git clone https://github.com/OmkarSahu0375/medbot.git

Go to the project folder

cd medbot

Create a virtual environment

python -m venv .venv

Activate the virtual environment

Windows

.venv\Scripts\activate

Linux / macOS

source .venv/bin/activate

Install the required packages

pip install -r requirements.txt

uv sync

🔑 Environment Variables

Create a .env file in the project root and add the following variables:

PINECONE_API_KEY=your_pinecone_api_key
GROQ_API_KEY=your_groq_api_key

Variable	Description
`PINECONE_API_KEY`	API key for Pinecone Vector Database
`GROQ_API_KEY`	API key for Groq LLM

▶️ Usage

Run the application:

uv run streamlit run streamlit_app.py

After the server starts, open the URL shown in the terminal (usually http://localhost:8501).

You can now ask medical questions and chat with MedBot.

⚠️ Limitations

This project is built for learning and demonstration purposes.
It only answers questions using the information available in the medical knowledge base.
If the required information is not available in the documents, the answer may be incomplete or incorrect.
AI-generated responses should always be verified with trusted medical sources.

📌 Disclaimer

This project is not a substitute for professional medical advice, diagnosis, or treatment. Always consult a qualified healthcare professional for medical concerns.