A feasibility study on an automated method to generate patient-specific dose distributions for radiotherapy using deep learning.

Journal: Medical physics
Published Date: Nov 23, 2018

Abstract

PURPOSE: To develop a method for predicting optimal dose distributions, given the planning image and segmented anatomy, by applying deep learning techniques to a database of previously optimized and approved Intensity-modulated radiation therapy treatment plans.

Authors

  • Xinyuan Chen
    National Cancer Center/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100021, China.
  • Kuo Men
    State Key Laboratory of Advanced Materials for Smart Sensing, GRINM Group Co., Ltd., Beijing 100088, China.
  • Yexiong Li
    Department of Radiation Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100021, China.
  • Junlin Yi
    Department of Radiation Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100021, China.
  • Jianrong Dai
    National Cancer Center/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China.

Keywords

External Resources

Popular Topics
Recent Journals