Deep learning based brain MRI registration driven by local-signed-distance fields of segmentation maps.

Journal: Medical physics
Published Date:

Abstract

BACKGROUND: Deep learning based unsupervised registration utilizes the intensity information to align images. To avoid the influence of intensity variation and improve the registration accuracy, unsupervised and weakly-supervised registration are combined, namely, dually-supervised registration. However, the estimated dense deformation fields (DDFs) will focus on the edges among adjacent tissues when the segmentation labels are directly used to drive the registration progress, which will decrease the plausibility of brain MRI registration.

Authors

  • Yue Yang
    Department of Nephrology, China-Japan Friendship Hospital, Beijing 100029, China.
  • Shunbo Hu
    School of Information, Linyi University, Linyi, 276005, China.
  • Lintao Zhang
    School of Information Science and Engineering, Linyi University, Linyi, Shandong, China.
  • Dinggang Shen
    School of Biomedical Engineering, ShanghaiTech University, Shanghai, China.