Deep Learning with Skip Connection Attention for Choroid Layer Segmentation in OCT Images.

Journal: Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference
Published Date:

Abstract

Since the thickness and shape of the choroid layer are indicators for the diagnosis of several ophthalmic diseases, the choroid layer segmentation is an important task. There exist many challenges in segmentation of the choroid layer. In this paper, in view of the lack of context information due to the ambiguous boundaries, and the subsequent inconsistent predictions of the same category targets ascribed to the lack of context information or the large regions, a novel Skip Connection Attention (SCA) module which is integrated into the U-Shape architecture is proposed to improve the precision of choroid layer segmentation in Optical Coherence Tomography (OCT) images. The main function of the SCA module is to capture the global context in the highest level to provide the decoder with stage-by-stage guidance, to extract more context information and generate more consistent predictions for the same class targets. By integrating the SCA module into the U-Net and CE-Net, we show that the module improves the accuracy of the choroid layer segmentation.

Authors

  • Xiaoqian Mao
    School of Electrical Engineering and Automation, Tianjin University, Tianjin 300072, China.
  • Yitian Zhao
  • Bang Chen
    Key Laboratory of Synthetic and Natural Functional Molecule Chemistry of Ministry of Education, College of Chemistry and Materials Science, National Demonstration Center for Experimental Chemistry Education, Northwest University, Xi'an, Shaanxi 710127, China.
  • Yuhui Ma
    Department of Computer and Communication Engineering, China University of Petroleum, Qingdao 266580, China.
  • Zaiwang Gu
  • Shenshen Gu
  • Jianlong Yang
    Cixi Institute of Biomedical Engineering, Ningbo Institute of Materials Technology and Engineering, Chinese Academy of Sciences, Ningbo, Zhejiang, China.
  • Jun Cheng
    School of Electrical and Information Technology, Yunnan Minzu University, Kunming, Yunnan 650500, PR China. Electronic address: jcheng6819@126.com.
  • Jiang Liu
    Department of Pharmacy, The Fourth Hospital of Hebei Medical University Shijiazhuang 050000, Hebei, China.