Variable generalization performance of a deep learning model to detect pneumonia in chest radiographs: A cross-sectional study.

Journal: PLoS medicine
Published Date: Nov 6, 2018

Abstract

BACKGROUND: There is interest in using convolutional neural networks (CNNs) to analyze medical imaging to provide computer-aided diagnosis (CAD). Recent work has suggested that image classification CNNs may not generalize to new data as well as previously believed. We assessed how well CNNs generalized across three hospital systems for a simulated pneumonia screening task.

Authors

  • John R Zech
    Department of Medicine, California Pacific Medical Center, San Francisco, California, United States of America.
  • Marcus A Badgeley
    Verily Life Sciences, South San Francisco, California, United States of America.
  • Manway Liu
    Verily Life Sciences, South San Francisco, California, United States of America.
  • Anthony B Costa
    Department of Neurological Surgery, Icahn School of Medicine, New York, New York, United States of America.
  • Joseph J Titano
    Department of Radiology, Icahn School of Medicine, New York, NY, USA.
  • Eric Karl Oermann
    Department of Neurosurgery, Icahn School of Medicine at Mount Sinai, New York City, NY, USA.

Keywords

External Resources

Popular Topics
Recent Journals