Researchers Evaluate Performance Of A Large Language Model In Phenotyping Postpartum Hemorrhage Patients

In a recent study published in npj Digital Medicine, researchers evaluated the performance of a large language model (LLM) in phenotyping postpartum hemorrhage (PPH) patients using discharge notes. The study developed an interpretable approach for phenotyping and subtyping of PPH cases by using the Flan-T5 LLM. The team identified over 138,000 individuals with an obstetric encounter at the Mass General Brigham hospitals in Boston between 1998 and 2015. Discharge summaries were used for NLP-based phenotyping and the team developed 24 PPH-related concepts and identified them in discharge notes by prompting the Flan-T5 model for two types of tasks – binary classification and text extraction.