A customizable deep learning model for nosocomial risk prediction from critical care notes with indirect supervision.

Journal: Journal of the American Medical Informatics Association : JAMIA

PMID: 32065628

Abstract

OBJECTIVE: Reliable longitudinal risk prediction for hospitalized patients is needed to provide quality care. Our goal is to develop a generalizable model capable of leveraging clinical notes to predict healthcare-associated diseases 24-96 hours in advance.

Authors

Travis R Goodwin

The University of Texas at Dallas, Richardson, TX.
Dina Demner-Fushman

Lister Hill National Center for Biomedical Communications, National Library of Medicine, Bethesda, MD.

Keywords

Acute Kidney Injury Anemia Area Under Curve Critical Care Deep Learning Electronic Health Records Humans Iatrogenic Disease Natural Language Processing Pressure Ulcer Retrospective Studies Risk Assessment ROC Curve Support Vector Machine

External Resources

View on PubMed Access via DOI PubMed (32065628)

A customizable deep learning model for nosocomial risk prediction from critical care notes with indirect supervision.

Abstract

Authors

Keywords

External Resources

Popular Topics

Recent Journals