ChatGPT-4 extraction of heart failure symptoms and signs from electronic health records.

Journal: Progress in cardiovascular diseases
PMID: 39442600

Abstract

BACKGROUND: Natural language processing (NLP) can facilitate research utilizing data from electronic health records (EHRs). Large language models can potentially improve NLP applications leveraging EHR notes. The objective of this study was to assess the performance of zero-shot learning using Chat Generative Pre-trained Transformer 4 (ChatGPT-4) for extraction of symptoms and signs, and compare its performance to baseline machine learning and rule-based methods developed using annotated data.

Authors

  • T Elizabeth Workman
    VA Salt Lake City Health Care System, Salt Lake City, Utah, USA.
  • Ali Ahmed
    Veterans Affairs Medical Center, Washington, DC; George Washington University, Washington, DC; Georgetown University, Washington, DC. Electronic address: ali.ahmed@va.gov.
  • Helen M Sheriff
    Veterans Affairs Medical Center, Washington, DC; George Washington University, Washington, DC.
  • Venkatesh K Raman
    Veterans Affairs Medical Center, Washington, DC; Georgetown University, Washington, DC.
  • Sijian Zhang
    Veterans Affairs Medical Center, Washington, DC.
  • Yijun Shao
    Veterans Affairs Medical Center, Washington, DC; George Washington University, Washington, DC.
  • Charles Faselis
    Veterans Affairs Medical Center, Washington, DC; George Washington University, Washington, DC; Uniformed Services University, Washington, DC.
  • Gregg C Fonarow
    Ahmanson-UCLA Cardiomyopathy Center, Division of Cardiology, Ronald Reagan-UCLA Medical Center, Los Angeles, California.
  • Qing Zeng-Treitler
    Veterans Affairs Medical Center, Washington, DC; George Washington University, Washington, DC.

Keywords

External Resources

Popular Topics
Recent Journals