Journal of Computer Applications ›› 2020, Vol. 40 ›› Issue (11): 3340-3345.DOI: 10.11772/j.issn.1001-9081.2020030390

• Virtual reality and multimedia computing • Previous Articles     Next Articles

Magnetic resonance image segmentation of articular synovium based on improved U-Net

WEI Xiaona1, XING Jiaqi2, WANG Zhenyu1, WANG Yingshan1, SHI Jie3, ZHAO Di4, WANG Hongzhi1   

  1. 1. Shanghai Key Laboratory of Magnetic Resonance(East China Normal University), Shanghai 200062, China;
    2. School of Acupuncture-Moxibustion and Tuina, Shanghai University of Traditional Chinese Medicine, Shanghai 200032, China;
    3. Shanghai GuangHua Hospital of Integrated Traditional Chinese and Western Medicine, Shanghai 200052, China;
    4. Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China
  • Received:2020-03-31 Revised:2020-05-26 Online:2020-11-10 Published:2020-06-03
  • Supported by:
    This work is partially supported by the Enterprise Horizontal Project of Shanghai Niumag Electronic Technology Company Limited (2017KFR0107).

基于改进U-Net的关节滑膜磁共振图像的分割

魏小娜1, 邢嘉祺2, 王振宇1, 王颖珊1, 石洁3, 赵地4, 汪红志1   

  1. 1. 上海市磁共振重点实验室(华东师范大学), 上海 200062;
    2. 上海市中医药大学 针灸推拿学院, 上海 200032;
    3. 上海市光华中西医结合医院, 上海 200052;
    4. 中国科学院 计算技术研究所, 北京 100190
  • 通讯作者: 汪红志(1975-),男,湖北黄冈人,高级工程师,博士,主要研究方向:磁共振(成像)技术、医学影像分析;hzwang@phy.ecnu.edu.cn
  • 作者简介:魏小娜(1990-),女,安徽宿州人,硕士研究生,主要研究方向:深度学习、医学图像处理;邢嘉祺(1996-),男,上海人,硕士研究生,主要研究方向:生物医学信号处理;王振宇(1996-),男,江西抚州人,硕士研究生,主要研究方向:深度学习、医学图像处理;王颖珊(1997-),女,浙江绍兴人,硕士研究生,主要研究方向:深度学习、医学图像处理;石洁(1976-),女,河北邯郸人,主任医师,硕士,主要研究方向:强脊类风湿关节炎的影像诊断;赵地(1978-),男,湖南长沙人,副教授,博士,主要研究方向:类脑计算
  • 基金资助:
    上海纽迈电子科技有限公司企业横向项目(2017KFR0107)。

Abstract: In order to accurately diagnose the synovitis patient's condition, doctors mainly rely on manual labeling and outlining method to extract synovial hyperplasia areas in the Magnetic Resonance Image (MRI). This method is time-consuming and inefficient, has certain subjectivity and is of low utilization rate of image information. To solve this problem, a new articular synovium segmentation algorithm, named 2D ResU-net segmentation algorithm was proposed. Firstly, the two-layer residual block in the Residual Network (ResNet) was integrated into the U-Net to construct the 2D ResU-net. Secondly, the sample dataset was divided into training set and testing set, and data augmentation was performed to the training set. Finally, all the training samples after augmentation were applied to the training of the network model. In order to test the segmentation effect of the model, the tomographic images containing synovitis in the testing set were selected for segmentation test. The final average segmentation accuracy indexes are as follow:Dice Similarity Coefficient (DSC) of 69.98%, IOU (Intersection over Union) index of 79.90% and Volumetric Overlap Error (VOE)of 12.11%. Compared with U-Net algorithm, 2D ResU-net algorithm has the DSC increased by 10.72%, IOU index increased by 4.24% and VOE decreased by 11.57%. Experimental results show that this algorithm can achieve better segmentation effect of synovial hyperplasia areas in MRI images, and can assist doctors to make diagnosis of the disease condition in time.

Key words: synovitis, magnetic resonance image, medical image segmentation, data augmentation, U-Net

摘要: 为了准确诊断滑膜炎患者病情,医生主要依靠手工标注和勾画的方法来提取磁共振图像(MRI)中的滑膜增生区域,该方法耗时长、效率低,具有一定的主观性且图像信息利用率低。针对这一问题,提出了一种新的关节滑膜分割算法,即2D ResU-net分割算法。首先,将残差网络(ResNet)中的两层结构的残差块融入到U-Net中,构建2D ResU-net;然后,将样本数据集分为训练集和测试集,而后对训练集进行数据增广;最后,将增广后的所有训练样本用于网络模型的训练。为了检测模型的分割效果,选取测试集中含滑膜炎的断层图像进行分割测试,最终平均分割精度指标可达到:Dice相似系数(DSC)69.98%,交并比(IOU)指标79.90%,体积重叠误差(VOE)系数12.11%。与U-Net算法相比,2D ResU-net算法的DSC系数提升了10.72%,IOU指标升高了4.24%,VOE系数降低了11.57%。实验结果表明,该算法对于MRI图像中的滑膜增生区域可以实现较好的分割效果,能够辅助医生对病情做出及时诊断。

关键词: 滑膜炎, 磁共振图像, 医学图像分割, 数据增广, U-Net

CLC Number: