基于深度监督残差网络的肝脏及肝肿瘤分割
基于深度监督残差网络的肝脏及肝肿瘤分割
作者:
作者单位:

作者简介:

通讯作者:

基金项目:

解放军总医院医疗大数据与人工智能研发基金项目(2019MBD-058, 2018MBD-005)

伦理声明:



The Liver and Liver Tumor Segmentation Based on Deeply Supervised Residual Unet
Author:
Ethical statement:

Affiliation:

Funding:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
    摘要:

    针对医生手动对肝脏肿瘤 CT 图像分割耗时、耗力,且易受主观判断影响的问题,该研究提 出一种深度监督残差网络(Deeply Supervised Residual Unet,DS-ResUnet)算法,以实现对腹部增强 CT 图像中肝脏及肝脏肿瘤区域进行全自动分割的目的。首先,利用公开发布的 MICCAI2017 肝脏肿瘤分 割(LiTS)挑战赛数据集,并使用 python 及 TensorFlow 开源框架进行数据分析;然后,构建深度监督 残差网络对肝脏及肝肿瘤图像进行自动分割;最后,通过平均 Dice 系数、全局 Dice 系数、Jaccard 系 数、平均对称表面距离(ASSD)、95% 豪斯多夫距离(HD95)、准确率和召回率七个评价指标对所提出 算法与 Unet 模型的性能进行比较分析。结果显示,所提出的 DS-ResUnet 算法在肝脏分割上的七个评 价指标结果依次为 96.06%、95.08%、92.54%、1.98 mm、12.87 mm、96.11%、96.06%,优于 Unet 模 型的结果(95.71%、94.52%、91.91%、2.41 mm、14.21 mm、95.48%、96.01%);在肝肿瘤分割上的 七个评价指标结果依次为 67.51%、76.65%、54.21%、6.65 mm、25.34 mm、80.39%、64.27%,也优 于 Unet 模型的结果(60.67%、73.47%、47.39%、9.43 mm、39.38 mm、79.61%、58.01%)。这表明所 提出的算法有效地提高了分割效果,实现了从 3D 腹部增强 CT 图像中全自动分割肝脏和肝肿瘤区域 的目的。

    Abstract:

    For the problem that doctors manually segment the liver tumor from CT image is time-consuming, labor-intensive, and susceptible to subjective judgment, we propose a deeply supervised residual Unet (DSResUnet) that incorporates residual link and deep supervision into Unet for more precise segmentation. The proposed method was evaluated on the public MICCAI 2017 liver segmentation (LiTS) challenge dataset with Dice coefficient, Jaccard coefficient, average symmetrical surface distance (ASSD), 95% Hausdorff distance (HD95), precision and recall. The experimental results show that the results on the above 7 evaluation indicators of liver segmentation with the proposed DS-ResUnet are 96.06%, 95.08%, 92.54%, 1.98 mm, 12.87 mm, 96.11%, and 96.06%, respectively, achieve superior results on almost all metrics to the widely-used Unet (95.71%, 94.52%, 91.91%, 2.41 mm, 14.21 mm, 95.48%, 96.01%). The results on the above 7 evaluation indicators of liver tumor segmentation with the proposed DS-ResUnet are 67.51%, 76.65%, 54.21%, 6.65 mm, 25.34 mm, 80.39%, and 64.27%, respectively, also better than that of the Unet (60.67%, 73.47%, 47.39%, 9.43 mm, 39.38 mm, 79.61%, 58.01%). Therefore, the proposed DS-ResUnet improves the segmentation results and achieves automatic segmentation of liver and liver tumor regions from the 3D abdominal enhanced CT image.

    参考文献
    相似文献
    引证文献
引用本文

引文格式
张家兵,张 耀,徐洪丽,沈舒宁,王 冬,刘同波,刘 坤,王彬华.基于深度监督残差网络的肝脏及肝肿瘤分割 [J].集成技术,2020,9(3):66-74

Citing format
ZHANG Jiabing, ZHANG Yao, XU Hongli, SHEN Shuning, WNAG Dong, LIU Tongbo, LIU Kun, WANG Binhua. The Liver and Liver Tumor Segmentation Based on Deeply Supervised Residual Unet[J]. Journal of Integration Technology,2020,9(3):66-74

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
历史
  • 收稿日期:2020-03-19
  • 最后修改日期:2020-03-30
  • 录用日期:
  • 在线发布日期: 2020-05-18
  • 出版日期: