[1]刘晶,徐皓,崔欣欣,等.基于多尺度边缘分割与混合注意力机制的脊柱CT图像分割[J].中国医学物理学杂志,2024,41(4):463-471.[doi:DOI:10.3969/j.issn.1005-202X.2024.04.011]
　LIU Jing,XU Hao,CUI Xinxin,et al.Spine CT image segmentation based on multi-scale boundary segmentation and hybrid attention mechanism[J].Chinese Journal of Medical Physics,2024,41(4):463-471.[doi:DOI:10.3969/j.issn.1005-202X.2024.04.011]
点击复制

基于多尺度边缘分割与混合注意力机制的脊柱CT图像分割()
分享到：

《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]

卷:
41卷
期数:
2024年第4期
页码:
463-471
栏目:
医学影像物理
出版日期:
2024-04-25

文章信息/Info

Title:
Spine CT image segmentation based on multi-scale boundary segmentation and hybrid attention mechanism
文章编号:
1005-202X（2024）04-0463-09
作者:
刘晶1; 徐皓1; 崔欣欣1; 田振宇1; 杨建兰2
1.甘肃中医药大学信息工程学院， 甘肃 兰州 730000； 2.泉州市正骨医院， 福建 泉州 362019
Author(s):
LIU Jing1;  XU Hao1;  CUI Xinxin1;  TIAN Zhenyu1;  YANG Jianlan2
1. School of Information Engineering, Gansu University of Chinese Medicine, Lanzhou 730000, China 2. Quanzhou Orthopedic-Traumatological Hospital, Quanzhou 362019, China
关键词:
脊柱分割; 3D U-Net; 椎骨边缘分割; 混合注意力机制
Keywords:
Keywords: spine segmentation 3D U-Net vertebral boundary segmentation hybrid attention mechanism
分类号:
R318；TP391.4
DOI:
DOI:10.3969/j.issn.1005-202X.2024.04.011
文献标志码:
A
摘要:
脊柱疾病的前期主要通过计算机断层扫描技术进行筛查与初步判断。为解决脊柱CT图像目前存在的椎骨结构复杂、分割精度不足等问题，提出一种基于3D U-Net框架的脊柱CT图像改进分割网络，通过融合SE残差单元、椎骨边缘分割模型与改进混合通道-空间注意力机制，在VerSe 19、VerSe 20与CTSpine1K脊柱数据集上进行分割训练与测试。多次测试实验结果表明，本文模型在保证分割精度和分割效率有效提高的同时具有较好的泛化性与鲁棒性，在Dice相似系数、豪斯多夫距离与平均表面距离上相较于其他先进网络分割精度更高。本文模型在现有脊柱分割的网络中具有更强的分割性能，可为放射科医生提供有效临床信息。
Abstract:
The early diagnosis of spinal diseases is mainly screened and initially diagnosed through computed tomography (CT). In view of the complex structure of vertebral bones and low segmentation accuracy in spinal CT images, a spinal CT image segmentation network based on 3D U-Net framework is proposed. The network which integrates squeeze-and-excitation residual module, vertebral boundary segmentation model, and improved hybrid channel-spatial attention mechanism is trained and tested on VerSe 19, VerSe 20, and CTSpine1K spinal datasets. Multiple experiments indicate that the model can effectively improve segmentation accuracy and efficiency while demonstrating good generalization and robustness. Compared with other advanced network models, the proposed network achieves higher segmentation accuracy in terms of Dice similarity coefficient, Hausdorff distance, and average symmetric surface distance. The proposed model exhibits superior segmentation performance among the existing spinal segmentation networks, providing radiologists with valuable clinical information.

相似文献/References:

[1]沈镇炯,彭昭,孟祥银,等.基于级联3D U-Net的CT和MR视交叉自动分割方法[J].中国医学物理学杂志,2021,38(8):950.[doi:DOI:10.3969/j.issn.1005-202X.2021.08.006]
　SHEN Zhenjiong,PENG Zhao,MENG Xiangyin,et al.Automatic optic chiasm segmentation using CT and MRI based on cascaded 3D U-Net[J].Chinese Journal of Medical Physics,2021,38(4):950.[doi:DOI:10.3969/j.issn.1005-202X.2021.08.006]
[2]温帆,杨萍,张鑫,等.基于特征增强的多分支U-Net肺结节分割[J].中国医学物理学杂志,2023,40(11):1343.[doi:DOI:10.3969/j.issn.1005-202X.2023.11.005]
　WEN Fan,YANG Ping,ZHANG Xin,et al.Pulmonary nodule segmentation using multi-branch U-Net based on feature enhancement[J].Chinese Journal of Medical Physics,2023,40(4):1343.[doi:DOI:10.3969/j.issn.1005-202X.2023.11.005]

备注/Memo

备注/Memo:
【收稿日期】2023-12-08 【作者简介】刘晶，硕士研究生，研究方向：医学图像处理，E-mail: 2442204612@qq.com 【通信作者】杨建兰，硕士生导师，副教授，研究方向：医学影像识别与应用，E-mail: FJYJL@gszy.edu.cn
更新日期/Last Update: 2024-04-25