Confidence-linked and uncertainty-based staged framework for phenotype validation using large language models.

Journal: Journal of the American Medical Informatics Association : JAMIA
Published Date: Aug 1, 2025

Abstract

OBJECTIVES: This study develops and validates the confidence-linked and uncertainty-based staged (CLUES) framework by integrating large language models (LLMs) with uncertainty quantification to assist manual chart review while ensuring reliability through a selective human review.

Authors

  • Sumin Lee
    Tomocube, Inc.
  • Hyeok-Hee Lee
    Institute for Innovation in Digital Healthcare, Yonsei University, Seoul 03722, Korea.
  • Hokyou Lee
    Institute for Innovation in Digital Healthcare, Yonsei University, Seoul 03722, Korea.
  • Kyu Sun Yum
    Department of Neurology, Chungbuk National University Hospital and Chungbuk National University College of Medicine, Cheongju-si, 28644, Chungcheongbuk-do, Republic of Korea. alcest@gmail.com.
  • Jang-Hyun Baek
    Department of Neurology, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul 03181, Korea.
  • Jaewon Khil
    Department of Preventive Medicine, Yonsei University College of Medicine, Seoul 03722, Korea.
  • Jaeyong Lee
    Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA.
  • Sojung Shin
    Department of Public Health, Yonsei University Graduate School, Seoul 03722, Korea.
  • Minsung Cho
    Department of Public Health, Yonsei University Graduate School, Seoul 03722, Korea.
  • Na Yeon Ahn
    Department of Public Health, Yonsei University Graduate School, Seoul 03722, Korea.
  • Seng Chan You
    Department of Biomedical Informatics, Ajou University School of Medicine, Suwon, South Korea.
  • Hyeon Chang Kim
    Division of Cardiology, Severance Cardiovascular Hospital, Yonsei University College of Medicine, Seoul, Korea.

Keywords

External Resources

Popular Topics
Recent Journals