Deep-learning-based image registration and automatic segmentation of organs-at-risk in cone-beam CT scans from high-dose radiation treatment of pancreatic cancer.

Journal: Medical physics
Published Date:

Abstract

PURPOSE: Accurate deformable registration between computed tomography (CT) and cone-beam CT (CBCT) images of pancreatic cancer patients treated with high biologically effective radiation doses is essential to assess changes in organ-at-risk (OAR) locations and shapes and to compute delivered dose. This study describes the development and evaluation of a deep-learning (DL) registration model to predict OAR segmentations on the CBCT derived from segmentations on the planning CT.

Authors

  • Xu Han
  • Jun Hong
  • Marsha Reyngold
    Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, 10065, USA.
  • Christopher Crane
    Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, 10065, USA.
  • John Cuaron
    Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, 10065, USA.
  • Carla Hajj
    Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, 10065, USA.
  • Justin Mann
    Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, 10065, USA.
  • Melissa Zinovoy
    Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, 10065, USA.
  • Hastings Greer
    Department of Computer Science, University of North Carolina, Chapel Hill, NC, 27599, USA.
  • Ellen Yorke
    Department of Medical Physics, Memorial Sloan-Kettering Cancer Center, New York, NY, 10065, USA.
  • Gig Mageras
    Department of Medical Physics, Memorial Sloan-Kettering Cancer Center, New York, USA; Department of Radiation Oncology, Memorial Sloan-Kettering Cancer Center, New York, USA.
  • Marc Niethammer
    Department of Computer Science, University of North Carolina, Chapel Hill, NC, 27599, USA.