A multimodal chatbot trained on medical anatomy books, integrating text, tables, and images within a vector database. It features a responsive Q&A interface supporting image-based searches to find similar images and related content. Powered by a multimodal LLM with Retrieval-Augmented Generation (RAG), it employs LangChain for interactions, Pinecone for vector search, and MongoDB for metadata storage.
https://anato-mind.streamlit.app/