Generating anthropomorphic phantoms using fully unsupervised deformable image registration with convolutional neural networks.

Journal: Medical physics
Published Date: Nov 9, 2020

Abstract

PURPOSE: Computerized phantoms have been widely used in nuclear medicine imaging for imaging system optimization and validation. Although the existing computerized phantoms can model anatomical variations through organ and phantom scaling, they do not provide a way to fully reproduce the anatomical variations and details seen in humans. In this work, we present a novel registration-based method for creating highly anatomically detailed computerized phantoms. We experimentally show substantially improved image similarity of the generated phantom to a patient image.

Authors

  • Junyu Chen
    Department of Electrical and Computer Engineering, Johns Hopkins University, Baltimore, MD, 21287, USA.
  • Ye Li
    Environment and Plant Protection Institute, Chinese Academy of Tropical Agricultural Science, Haikou 571010, People's Republic of China; Key Laboratory of Monitoring and Control of Tropical Agricultural and Forest Invasive Alien Pests, Ministry of Agriculture, Haikou 571010, People's Republic of China.
  • Yong Du
    Biomedical Engineering DepartmentUniversity of Houston Houston TX 77204 USA.
  • Eric C Frey
    Department of Electrical and Computer Engineering, Johns Hopkins University, Baltimore, MD, 21287, USA.

Keywords

External Resources

Popular Topics
Recent Journals