Clinical applications of machine learning in the diagnosis, classification, and prediction of heart failure.

Journal: American heart journal
Published Date:

Abstract

Machine learning and artificial intelligence are generating significant attention in the scientific community and media. Such algorithms have great potential in medicine for personalizing and improving patient care, including in the diagnosis and management of heart failure. Many physicians are familiar with these terms and the excitement surrounding them, but many are unfamiliar with the basics of these algorithms and how they are applied to medicine. Within heart failure research, current applications of machine learning include creating new approaches to diagnosis, classifying patients into novel phenotypic groups, and improving prediction capabilities. In this paper, we provide an overview of machine learning targeted for the practicing clinician and evaluate current applications of machine learning in the diagnosis, classification, and prediction of heart failure.

Authors

  • Cameron R Olsen
    Division of Cardiology, Duke University Medical Center, Durham, NC. Electronic address: cameron.olsen@duke.edu.
  • Robert J Mentz
    Division of Cardiology, Duke University Medical Center, Durham, NC.
  • Kevin J Anstrom
    Department of Biostatistics and Bioinformatics, Duke University, Durham, NC.
  • David Page
    Duke University, Department of Biostatistics and Bioinformatics, Durham, NC, USA.
  • Priyesh A Patel
    Sanger Heart and Vascular Institute, Atrium Health, Charlotte, NC.