《计算机应用》唯一官方网站 ›› 2023, Vol. 43 ›› Issue (7): 2303-2310.DOI: 10.11772/j.issn.1001-9081.2022060803

• 多媒体计算与计算机仿真 • 上一篇    

基于多尺度特征融合和网格注意力机制的三维肝脏影像分割方法

郑帅1,2,3, 张晓龙1,2,3(), 邓鹤1,2,3, 任宏伟4   

  1. 1.武汉科技大学 计算机科学与技术学院, 武汉 430065
    2.武汉科技大学 大数据科学与工程研究院, 武汉 430065
    3.智能信息处理与实时工业系统湖北省重点实验室(武汉科技大学), 武汉 430065
    4.武汉科技大学附属天佑医院, 武汉 430064
  • 收稿日期:2022-06-06 修回日期:2022-08-04 接受日期:2022-08-11 发布日期:2022-08-26 出版日期:2023-07-10
  • 通讯作者: 张晓龙
  • 作者简介:郑帅(1998—),男,湖北孝感人,硕士研究生,主要研究方向:计算机视觉、深度学习;
    张晓龙(1963—),男,江西吉安人,教授,博士,主要研究方向:人工智能、机器学习、数据挖掘、生物信息处理;
    邓鹤(1977—),湖北孝感人,教授,博士,主要研究方向:深度学习、图像处理;
    任宏伟(1978—),女,河北滦城人,副主任医师,硕士,主要研究方向:消化系统、中枢神经系统及骨肌系统的影像学诊断。
  • 基金资助:
    国家自然科学基金资助项目(61972299)

3D liver image segmentation method based on multi-scale feature fusion and grid attention mechanism

Shuai ZHENG1,2,3, Xiaolong ZHANG1,2,3(), He DENG1,2,3, Hongwei REN4   

  1. 1.School of Computer Science and Technology,Wuhan University of Science and Technology,Wuhan Hubei 430065,China
    2.Institute of Big Data Science and Engineering,Wuhan University of Science and Technology,Wuhan Hubei 430065,China
    3.Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System (Wuhan University of Science and Technology),Wuhan Hubei 430065,China
    4.Tianyou Hospital Affiliated to Wuhan University of Science and Technology,Wuhan Hubei 430064,China
  • Received:2022-06-06 Revised:2022-08-04 Accepted:2022-08-11 Online:2022-08-26 Published:2023-07-10
  • Contact: Xiaolong ZHANG
  • About author:ZHENG Shuai, born in 1998, M. S. candidate. His research interests include computer vision, deep learning.
    ZHANG Xiaolong, born in 1963, Ph. D., professor. His research interests include artificial intelligence, machine learning, data mining, biological information processing.
    DENG He, born in 1977, Ph. D., professor. His research interests include deep learning, image processing.
    REN Hongwei, born in 1978, M. S., deputy chief physician. Her research interests include imaging diagnosis of digestive system, central nervous system and musculoskeletal system.
  • Supported by:
    National Natural Science Foundation of China(61972299)

摘要:

在计算机断层扫描(CT)和磁共振成像(MRI)的影像中肝脏与邻近脏器的灰度值相似性都比较高,为自动精确地分割肝脏,提出一种基于多尺度特征融合和网格注意力机制的三维肝脏影像分割方法MAGNet (Multi-scale feature fusion And Grid attention mechanism Network)。首先,通过注意力引导连接模块来连接高层特征和低层特征以提取出重要的上下文信息,并且在注意力引导连接模块中引入网格注意力机制来关注感兴趣的分割区域;然后,通过在单个特征图中按通道数进行分层连接形成多尺度特征融合模块,并用该模块替换基础卷积块以获取多尺度语义信息;最后,利用深度监督机制解决梯度消失、梯度爆炸和收敛过慢等问题。实验结果表明:在3DIRCADb数据集上,与U3-Net+DC方法相比,MAGNet在Dice相似系数(DSC)指标上提升了0.10个百分点,在相对体积差(RVD)指标上降低了1.97个百分点;在Sliver07数据集上,与CANet方法相比,MAGNet在DSC指标上提升了0.30个百分点,在体素重叠误差(VOE)指标上降低了0.68个百分点,在平均对称表面距离(ASD)和对称位置表面距离的均方根(RMSD)指标上分别降低了0.03 mm和0.22 mm;在某医院肝脏MRI数据集上,MAGNet在所有指标上也均具有良好的结果。另外,将MAGNet应用于3DIRCADb数据集和某医院肝脏MRI数据集进行混合形成的数据集,也取得了非常有竞争力的分割效果。

关键词: 三维肝脏医疗影像, 语义分割, 深度学习, 多尺度特征融合, 注意力机制

Abstract:

Due to the high similarity of gray values among liver and adjacent organs in Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) images, a 3D liver image segmentation method based on multi-scale feature fusion and grid attention mechanism, namely MAGNet (Multi-scale feature fusion And Grid attention mechanism Network), was proposed to segment liver automatically and accurately. Firstly, high-level features and low-level features were connected by the attention-guided concatenation module to extract important context information, and the grid attention mechanism was introduced in the attention-guided concatenation module to focus on the segmentation region of interest. Then, the multi-scale feature fusion module was formed by the layered connection in a single feature map according to the number of channels, and this module was used to replace the basic convolutional block to obtain multi-scale semantic information. Finally, the deep supervision mechanism was utilized to solve the problems of vanishing gradient, exploding gradient and slow convergence. Experimental results show that on 3DIRCADb dataset, compared with the U3-Net+DC method, MAGNet improves the Dice Similarity Coefficient (DSC) metric by 0.10 percentage points and reduces the Relative Volume Difference (RVD) metric by 1.97 percentage points; on Sliver07 dataset, compared with the CANet method, MAGNet improves the DSC metrics by 0.30 percentage points, reduces Volumetric Overlap Error (VOE) metrics by 0.68 percentage points, and reduces the Average Symmetric Surface Distance (ASD) and Root Mean Square Symmetric Surface Distance (RMSD) metrics 0.03 mm and 0.22 mm respectively; on the liver MRI dataset of a hospital, MAGNet also has good results on all metrics. Besides, MAGNet was applied to a mixed dataset of 3DIRCADb dataset and the hospital liver MRI dataset above, and a competitive segmentation result was also achieved.

Key words: 3D liver medical image, semantic segmentation, deep learning, multi-scale feature fusion, attention mechanism

中图分类号: