Journal of Computer Applications ›› 2020, Vol. 40 ›› Issue (8): 2392-2397.DOI: 10.11772/j.issn.1001-9081.2020030318

• Virtual reality and multimedia computing • Previous Articles     Next Articles

Rectal tumor segmentation method based on improved U-Net model

GAO Haijun1, ZENG Xiangyin2, PAN Dazhi1,3, ZHENG Bochuan1,3   

  1. 1. School of Mathematics&Information, China West Normal University, Nanchong Sichuan 637009, China;
    2. School of Computer Science, China West Normal University, Nanchong Sichuan 637009, China;
    3. Institute of Computing Method and Application Software, China West Normal University, Nanchong Sichuan 637009, China
  • Received:2020-03-19 Revised:2020-05-02 Online:2020-08-10 Published:2020-05-19
  • Supported by:
    This work is partially supported by the Sichuan Science and Technology Program (2019YFG0299), the Fundamental Research Funds of China West Normal University (19B045).


高海军1, 曾祥银2, 潘大志1,3, 郑伯川1,3   

  1. 1. 西华师范大学 数学与信息学院, 四川 南充 637009;
    2. 西华师范大学 计算机学院, 四川 南充 637009;
    3. 西华师范大学 计算方法与应用软件研究所, 四川 南充 637009
  • 通讯作者: 郑伯川(1974-),男,四川自贡人,教授,博士,CCF会员,主要研究方向:机器学习、深度学习、计算机视觉,
  • 作者简介:高海军(1992-),男,四川通江人,硕士研究生,CCF会员,主要研究方向:计算机视觉、深度学习、智能计算;曾祥银(1994-),男,四川大竹人,硕士研究生,主要研究方向:机器学习、深度学习;潘大志(1974-),男,四川三台人,教授,博士,CCF会员,主要研究方向:智能计算。
  • 基金资助:

Abstract: In the diagnosis of rectal cancer, if the rectal tumor area can be automatically and accurately segmented from Computed Tomography (CT) images, it will help doctors make a more accurate and rapid diagnosis. Aiming at the problem of rectal tumor segmentation, an automatic segmentation method of rectal tumor based on improved U-Net model was proposed. Firstly, the sub coding modules were embedded in the U-Net model encoder of different levels to improve the feature extraction ability of the model. Secondly, by comparing the optimization performances of different optimizers, the most suitable optimizer was determined to train the model. Finally, data augmentation was performed to the training set to make the model more fully trained, so as to improve the segmentation performance. Experimental results show that compared with U-Net, Y-Net and FocusNetAlpha network models, the segmentation region obtained by the improved model is closer to the real tumor region, and the segmentation performance of this model for small objects is more prominent; at the same time, the proposed model is superior to other three models on three evaluation indexes including precision, recall and Dice coefficient, which can effectively segment the rectal tumor area.

Key words: medical image segmentation, tumor segmentation, rectal tumor, Convolutional Neural Network (CNN), U-Net

摘要: 诊断直肠癌时,如果能够从CT图像中自动准确分割出直肠肿瘤区域,将有助于医生进行更准确和快速的诊断。针对直肠肿瘤分割问题,提出基于U-Net改进模型的直肠肿瘤自动分割方法。首先在U-Net模型的每级编码器中嵌入子编码模块提升模型特征提取能力;其次通过对比不同优化器的优化性能,获得最适合的优化器用于训练模型;最后对训练集进行数据扩充使模型得到更充分的训练,从而提高分割性能。与U-Net、Y-Net和FocusNetAlpha三种网络模型进行的对比实验表明:所提改进模型得到的分割区域与真实肿瘤区域更接近,对小目标的分割性能更突出,该模型的查准率、查全率和Dice系数三个评价指标都优于对比的模型,能有效分割直肠肿瘤区域。

关键词: 医学图像分割, 肿瘤分割, 直肠肿瘤, 卷积神经网络, U-Net

CLC Number: