MedCAT 2.0: LLM-Powered Medical Concept Annotation 🏥

Recorded October 25, 2024 at SNOMED CT Expo 2024 in Seoul, Korea.

202463 MedCAT 2.0 - Medical Concept Annotation Tool. Using large language models to extract SNOMED CT annotations from EHRs - Adam Sutton (UK)

Large Language Models (LLMs) fine-tuned to answer medical domain questions often lack transparency in the training process, leaving uncertainties about their capacity to address clinical queries. Our study employs the structured knowledge from the graphs from SNOMED CT to evaluate the performance of Large Language Models (LLMs) in healthcare. Our goal is to provide a comprehensive assessment of LLMs' performance in the healthcare industry.