MediBot

MediBot is a healthcare-focused AI chatbot that lets you query PDF documents such as clinical reports, medical papers, or patient records using Retrieval-Augmented Generation (RAG). Powered by LangChain, Groq LLM, HuggingFace embeddings, and ChromaDB, MediBot delivers fast, reliable, and context-aware answers directly from your documents.

---

Overview

MediBot transforms static medical PDFs into an interactive Q&A experience. Whether you're a healthcare professional, researcher, or student, you can upload documents and ask questions in natural language—getting grounded answers instantly.

It uses:

• Sentence embeddings via HuggingFace
• Fast inference through Groq LLM
• Vector-based semantic search to retrieve relevant content chunks

Demo: LinkedIn Demo

---

Features

• Upload and parse medical or general PDFs
• Automatic document chunking and indexing
• Semantic search with HuggingFace all-MiniLM-L12-v2
• Fast, context-aware answers using Groq LLM
• Simple and clean UI with Streamlit
• API key management via .env

---

Tech Stack

Component	Tool/Service
LLM	Groq API
Embeddings	HuggingFace
Framework	LangChain
Frontend	Streamlit
Vector DB	ChromaDB via LangChain

---

Setup

Clone the repo:

git clone https://github.com/Hassan123j/medibot.git
cd medibot

Install the required packages:

pip install -r requirements.txt

Set your Groq API key via environment variable:

export GROQ_API_KEY=your_api_key_here

Or add it to a .env file:

GROQ_API_KEY=your_api_key_here

---

Run the App

streamlit run app.py

This will launch the MediBot interface in your browser. Upload a PDF, ask a question, and receive medically grounded responses.

---

Project Structure

medibot/
├── app.py            # Main Streamlit app
├── requirements.txt  # Python dependencies
├── .env              # Optional: API key
└── README.md         # Project documentation

---

Contributing

We welcome contributions! Open an issue or submit a pull request with improvements, bug fixes, or suggestions.