Welcome to the repository for our capstone project, part of the Master of Science in Applied Data Science program at the University of Chicago. Our team is exploring the intersection of Generative Pre-trained Transformers (GPT) and healthcare, with a specific focus on implementing a Retrieval-Augmented Generation (RAG) framework.
Our project aims to develop a system that retrieves queried information from a multimodal vector space and generates coherent responses. This application is tailored for use with materials from UChicago Medicine, leveraging a curated database of textbooks, research papers, and other selected resources.
- Bruna Medeiros
- John Melel
- Kyler Rosen
- Samuel Martinez Koss
- Utilize GPT models within a RAG-based framework to provide accurate and meaningful answers to healthcare-related queries.
- Build a multimodal vector space to efficiently manage and retrieve data from diverse sources.
- Focus on curated, high-quality datasets including medical textbooks, research papers, and other academic materials.