Classifying Unstructured Text in Electronic Health Records for Mental Health Prediction Models: Large Language Model Evaluation Study.

Journal: JMIR medical informatics
PMID: 39864953

Abstract

BACKGROUND: Prediction models have demonstrated a range of applications across medicine, including using electronic health record (EHR) data to identify hospital readmission and mortality risk. Large language models (LLMs) can transform unstructured EHR text into structured features, which can then be integrated into statistical prediction models, ensuring that the results are both clinically meaningful and interpretable.

Authors

  • Nicholas C Cardamone
    Department of Psychiatry, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, United States.
  • Mark Olfson
    Department of Psychiatry, New York State Psychiatric Institute, Columbia University Medical Center, New York.
  • Timothy Schmutte
    Department of Psychiatry, Yale School of Medicine, New Haven, CT, United States.
  • Lyle Ungar
    University of Pennsylvania, USA.
  • Tony Liu
    University of Pennsylvania, USA.
  • Sara W Cullen
    School of Social Policy & Practice, University of Pennsylvania, Philadelphia, PA, United States.
  • Nathaniel J Williams
    School of Social Work, Boise State University, Boise, ID, United States.
  • Steven C Marcus
    School of Social Policy & Practice, University of Pennsylvania, Philadelphia, PA, United States.

Keywords

External Resources

Popular Topics
Recent Journals