Quantifying uncertainty in machine learning classifiers for medical imaging.

Journal: International journal of computer assisted radiology and surgery
Published Date: Mar 12, 2022

Abstract

PURPOSE: Machine learning (ML) models in medical imaging (MI) can be of great value in computer aided diagnostic systems, but little attention is given to the confidence (alternatively, uncertainty) of such models, which may have significant clinical implications. This paper applied, validated, and explored a technique for assessing uncertainty in convolutional neural networks (CNNs) in the context of MI.

Authors

  • John Valen
    Department of Medical Imaging, University of Toronto, Toronto, ON, M5T 1W7, Canada.
  • Indranil Balki
    Department of Medical Imaging, University of Toronto, Toronto, Ontario, Canada.
  • Mauro Mendez
    Department of Medical Imaging, University of Toronto, Toronto, ON, M5T 1W7, Canada.
  • Wendi Qu
    Department of Medical Imaging, University of Toronto, Toronto, ON, M5T 1W7, Canada.
  • Jacob Levman
    Division of Newborn Medicine, Department of Medicine, Boston Children's Hospital, Harvard Medical SchoolBoston, MA, USA; Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General HospitalCharlestown, MA, USA.
  • Alexander Bilbily
  • Pascal N Tyrrell
    Department of Medical Imaging, University of Toronto, Toronto, Ontario, Canada; Department of Statistical Sciences, University of Toronto, Toronto, Ontario, Canada. Electronic address: pascal.tyrrell@utoronto.ca.

Keywords

External Resources

Popular Topics
Recent Journals