Utility of Named Entity Recognition and Large Language Models in Cancer Research Informatics: Complimentary Application of Prodigy and LLAMA2

Poster Number: P86
Presentation Time: 05:00 PM - 06:30 PM

Abstract Keywords: Deep Learning, Large Language Models (LLMs), Natural Language Processing
Primary Track: Applications
Programmatic Theme: Clinical Research Informatics

Named Entity Recognition (NER) and Large Language Models (LLMs) can be used to extract critical cancer-related data from clinical notes. Using Prodigy and LLAMA2 platforms, we achieved high accuracy in identifying tumor biomarkers, cancer subtypes, and environmental exposures across several cancer types. NER and LLM approaches complement each other, when applied creatively to different use cases. Future enhancements include standardizing and integrating data outputs into broader research frameworks.

Speaker(s):
Ferris Hussein, Computer Science
NYU Langone Health

Author(s):
Rimma Belenkaya, MS, MA - NYU Langone; Jessica Sommer, BS - NYU Langone Health;