Summer Research Project Grant Report
Cross-disciplinary Domains: AI, Medical
Research Project Objective
The Paediatric Anaesthesia Research team at Perth Children’s Hospital requires a platform that allows efficient and effective organisation of clinical research studies. The team has the ability to host a large language model to run this platform. A requirement is to be able to flexibly configure available knowledge bases and tools to adapt to multiple different studies. This platform will need to be cohesive for many members to use and integrate AI features to run and administrate studies as well as offer observation tools for iterative development to continually improve the platform.
Background
In medical and clinical research, large volumes of structured and unstructured data must be managed. Accessing and understanding this information often requires reviewing extensive documentation, which can be time-consuming and inefficient. By utilising artificial intelligence, these processes can be simplified and sped up, enabling medical professionals to achieve better results. Retrieval Augmented Generation (RAG)1 is a technique used to guide large language models (LLMs) to generate responses based on specific knowledge bases. This is achieved by converting text into vector embeddings and comparing similarities between user queries and stored data, then returning the most similar data as relevant information.
