GatorTron AI Model

GatorTron™ is an artificial intelligence transformer natural language processing model intended to accelerate research and medical decision-making by extracting insights from massive volumes of clinical data with unprecedented speed and clarity. It is a foundation model that can be adopted in many clinical research and healthcare applications involving healthcare text analytics.

GatorTron models are encoder-only large language models (LLMs) developed using extensive amounts of all types of clinical text data accumulated in a recent decade at the University of Florida Health. GatorTron models can be applied to identify important patient information of interest (i.e., clinical concepts) from clinical text, identify the relations among clinical concepts, determine clinical phenotype groups, and support various predictive models for readmission, disease risk, and clinical decision support.