Ensuring electronic medical record simulation through better training, modeling, and evaluation.

Journal: Journal of the American Medical Informatics Association : JAMIA
PMID: 31592533

Abstract

OBJECTIVE: Electronic medical records (EMRs) can support medical research and discovery, but privacy risks limit the sharing of such data on a wide scale. Various approaches have been developed to mitigate risk, including record simulation via generative adversarial networks (GANs). While showing promise in certain application domains, GANs lack a principled approach for EMR data that induces subpar simulation. In this article, we improve EMR simulation through a novel pipeline that (1) enhances the learning model, (2) incorporates evaluation criteria for data utility that informs learning, and (3) refines the training process.

Authors

  • Ziqi Zhang
    Department of Electrical Engineering and Computer Science, Vanderbilt University, Nashville, Tennessee, USA.
  • Chao Yan
    School of Pharmacy, Shanghai Jiao Tong University, Shanghai 200240, China.
  • Diego A Mesa
    Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee, USA.
  • Jimeng Sun
    College of Computing Georgia Institute of Technology Atlanta, GA, USA.
  • Bradley A Malin
    Vanderbilt University, Nashville, TN.

Keywords

External Resources

Popular Topics
Recent Journals