Accurate colorectal tumor segmentation for CT scans based on the label assignment generative adversarial network.

Journal: Medical physics
Published Date: Jun 25, 2019

Abstract

PURPOSE: Colorectal tumor segmentation is an important step in the analysis and diagnosis of colorectal cancer. This task is a time consuming one since it is often performed manually by radiologists. This paper presents an automatic postprocessing module to refine the segmentation of deep networks. The label assignment generative adversarial network (LAGAN) is improved from the generative adversarial network (GAN) and assigns labels to the outputs of deep networks. We apply the LAGAN to segment colorectal tumors in computed tomography (CT) scans and explore the performances of different combinations of deep networks.

Authors

  • Xiaoming Liu
    College of Agriculture, Northeast Agricultural University, Harbin, China.
  • Shuxu Guo
    College of Electronic Science and Engineering, Jilin University, Changchun, China.
  • Huimao Zhang
    Department of Radiology, The First Hospital of Jilin University, No.1, Xinmin Street, Changchun 130021, China (Y.W., M.L., Z.M., J.W., K.H., Q.Y., L.Z., L.M., H.Z.). Electronic address: huimao@jlu.edu.cn.
  • Kan He
    Department of Radiology, The First Hospital of Jilin University, No.1, Xinmin Street, Changchun 130021, China (Y.W., M.L., Z.M., J.W., K.H., Q.Y., L.Z., L.M., H.Z.).
  • Shengnan Mu
    Department of Radiology, The First Hospital of Jilin University, Changchun, Jilin Province, 130021, China.
  • Yu Guo
    Animal Disease Control Center of Inner Mongolia, Hohhot, China.
  • Xueyan Li
    College of Electronic Science and Engineering, Jilin University, Changchun, China. Electronic address: leexy@jlu.edu.cn.

Keywords

External Resources

Popular Topics
Recent Journals