Annotated normal CT data of the abdomen for deep learning: Challenges and strategies for implementation.

Journal: Diagnostic and interventional imaging
Published Date:

Abstract

PURPOSE: The purpose of this study was to report procedures developed to annotate abdominal computed tomography (CT) images from subjects without pancreatic disease that will be used as the input for deep convolutional neural networks (DNN) for development of deep learning algorithms for automatic recognition of a normal pancreas.

Authors

  • S Park
  • L C Chu
    The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, School of Medicine, 601N. Caroline Street, Baltimore, MD 21287, USA.
  • E K Fishman
    The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, School of Medicine, 601N. Caroline Street, Baltimore, MD 21287, USA.
  • A L Yuille
    Department of Computer Science, Johns Hopkins University, School of Arts and Sciences, Baltimore, MD 21218, USA.
  • B Vogelstein
    Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University, School of Medicine, Baltimore, MD 21287, USA; Johns Hopkins University, School of Medicine, Ludwig Center for Cancer Genetics and Therapeutics, Baltimore, MD 21205, USA.
  • K W Kinzler
    Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University, School of Medicine, Baltimore, MD 21287, USA; Johns Hopkins University, School of Medicine, Ludwig Center for Cancer Genetics and Therapeutics, Baltimore, MD 21205, USA.
  • K M Horton
    The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, School of Medicine, 601N. Caroline Street, Baltimore, MD 21287, USA.
  • R H Hruban
    Department of Pathology, The Sol Goldman Pancreatic Cancer Research Center, Johns Hopkins University, School of Medicine, Baltimore, MD 21205, USA.
  • E S Zinreich
    The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, School of Medicine, 601N. Caroline Street, Baltimore, MD 21287, USA.
  • D F Fouladi
    The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, School of Medicine, 601N. Caroline Street, Baltimore, MD 21287, USA.
  • S Shayesteh
    The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, School of Medicine, 601N. Caroline Street, Baltimore, MD 21287, USA.
  • J Graves
    The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, School of Medicine, 601N. Caroline Street, Baltimore, MD 21287, USA.
  • S Kawamoto
    The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, School of Medicine, 601N. Caroline Street, Baltimore, MD 21287, USA. Electronic address: skawamo1@jhmi.edu.