Medical ChatBot

A web-based Medical ChatBot powered by FastAPI, LangChain, and Ollama LLM, providing accurate and up-to-date medical information. The chatbot retrieves information from MedlinePlus datasets and answers user queries in clear and concise language.

---

Features

• Ask questions about medical conditions such as diabetes, cancer, fever, influenza, depression, anemia, and obesity.
• Retrieval-based responses using FAISS vector store and embeddings.
• Clean, responsive frontend with a chat interface.
• Real-time chatbot interaction with loading feedback.
• Sections for About Us and Services.
• Encourages users to consult licensed healthcare providers for personal concerns.

---

Tech Stack

• Backend: FastAPI
• Frontend: HTML, CSS, JavaScript
• LLM: Ollama (llama2 model)
• Vector Store: FAISS
• Embeddings: sentence-transformers/all-mpnet-base-v2
• Data Source: MedlinePlus API
• Python Libraries:
  • requests
  • BeautifulSoup4
  • langchain
  • fastapi
  • pydantic

---

Usage

-Type a medical question in the input field.

-Click Send or press Enter.

-Wait for the chatbot to retrieve information and display the answer with sources.

-Explore the About Us and Services sections for additional information.

Notes

-The chatbot is retrieval-based, providing information only from MedlinePlus.

-Always consult a licensed healthcare provider for personal medical advice.

-The FAISS vector store will be built on first run and stored in the vector_store/ directory for faster queries next time.