[1]胡俊,李平.采用改进Snake模型的胸片肺野自动分割方法[J].华侨大学学报(自然科学版),2022,43(3):361-370.[doi:10.11830/ISSN.1000-5013.202012032]
　HU Jun,LI Ping.Automatic Segmentation Method of Lung Field on Chest Radiograph Using Improved Snake Model[J].Journal of Huaqiao University(Natural Science),2022,43(3):361-370.[doi:10.11830/ISSN.1000-5013.202012032]
点击复制

采用改进Snake模型的胸片肺野自动分割方法()
分享到：

《华侨大学学报（自然科学版）》[ISSN:1000-5013/CN:35-1079/N]

卷:
第43卷
期数:
2022年第3期
页码:
361-370
栏目:
出版日期:
2022-05-10

文章信息/Info

Title:
Automatic Segmentation Method of Lung Field on Chest Radiograph Using Improved Snake Model
文章编号:
1000-5013(2022)03-0361-10
作者:
胡俊;  李平
华侨大学 信息科学与工程学院, 福建 厦门 361021
Author(s):
HU Jun;  LI Ping
College of Information Science and Engineering, Huaqiao University, Xiamen 361021, China
关键词:
X光胸片;  Snake模型;  肺野分割;  Otsu法;  自动初始化
Keywords:
X-ray rabat;  snake model;  lung segmentation;  Otsu method;  automatic initialization
分类号:
R816.4;TP391.41
DOI:
10.11830/ISSN.1000-5013.202012032
文献标志码:
A
摘要:
为了解决传统Snake模型应用于X光胸片肺野分割时,对人工初始化轮廓的选择敏感、对凹陷区域分割不准确等问题,提出一种基于自动初始化Snake模型的X光胸片肺野自动分割方法.该方法首先通过Otsu法对原始图像进行二值化,得到包含肺野、背景区域的二值图像,并经过图像取反和连通域处理,运用形态学方法,得到只含有肺野区域的二值图像;然后,通过边界提取,完成对Snake模型轮廓的自动初始化;最后通过Snake模型的演化,得到分割结果.实验结果表明:该方法能摆脱Snake模型对人工初始化轮廓的依赖,提高分割的鲁棒性,同时对凹陷区域的分割更准确,具有更好的分割效果.
Abstract:
In order to solve the problems such as sensitive selection of artificial initial contour and inaccurate segmentation of the concave area when traditional Snake model is applied to segmentation of lung field in X-ray chest film, an automatic lung field segmentation method based on the automatic initialization Snake model was proposed. In this method, the original image was binarized by Otsu method to obtain binarization images including lung field and background area. After image inversion and connected domain processing, the binary image containing only lung field area was obtained by morphological method. Then, automatic initialization of Snake model contour can be completed through boundary extraction. Finally, the segmentation results are obtained through the evolution of Snake model. The experimental results show that this method can get rid of the dependence of Snake model on artificial initial contour, improve the robustness of segmentation, and at the same time, segment the sunken area more accurately and have better segmentation effect.

参考文献/References:

[1] BROWN M S,WILSON L S,DOUST B D,et al.Knowledge-based method for segmentation and analysis of lung boundaries in chest X-ray images[J].Computerized Medical Imaging and Graphics,1998,22(6):463-477.DOI:10.1016/S0895-6111(98)00051-2.
[2] 张文莉,吕晓琪,谷宇,等.基于肺部CT图像中肺实质分割的研究进展[J].中国医学物理学杂志,2017,34(9):902-907.DOI:10.3969/j.issn.1005-202X.2017.09.009.
[3] 曹新华,蒋大宗,郑崇勋.一种数字胸片图像肺部区域的自动分割方法[J].生物医学工程学杂志,1998,15(4):3-5.
[4] 张继武,张道兵,史舒娟,等.基于水平集方法的数字胸片图像分割[J].中国图象图形学报,2004,9(12):65-71.DOI:10.3969/j.issn.1006-8961.2004.12.009.
[5] 刘炎,邱天爽,郭冬梅.基于柔性形态学与聚类算法的X光胸片肺部分割[J].中国生物医学工程学报,2007,26(5):684-689.DOI:10.3969/j.issn.0258-8021.2007.05.009.
[6] CHEN H J,RUAN S J,HUANG S W,et al.Lung X-ray segmentation using deep convolutional neural networks on contrast-enhanced binarized images[J].Mathematics,2020,8(4):545-558.DOI:10.3390/math8040545.
[7] CHEN Bingzhi,ZHANG Zheng,LIN Jianyong,et al.Two-stream collaborative network for multi-label chest X-ray image classification with lung segmentation[J].Pattern Recognition Letters,2020,135(8):221-227.DOI:10.1016/j.patrec.2020.04.016.
[8] KASS M,WITKIN A,TERZOPOULOS D.Snakes: Active contour models[J].International Journal of Computer Vision,1988,1(4):321-331.DOI:10.1007/BF00133570.
[9] 石锐,黄向娟.基于改进Snake模型的肺部图像分割[J].计算机工程与应用,2014,50(15):164-167,186.DOI:10.3778/j.issn.1002-8331.1209-0026.
[10] 吴刚.数字X光胸片肺部分割的研究[D].深圳:中国科学院深圳先进技术研究院,2014.
[11] 朱伟,刘健,竺明月,等.基于高斯-拉普拉斯金字塔的DR图像增强改进算法研究[J].中国医疗器械杂志,2019,43(1):10-13.DOI:10.3969/j.issn.1671-7104.2019.01.003.
[12] 杨晓倩,毕惟红.改进snake模型的自动初始化轮廓检测方法[J].计算机应用研究,2020,37(增刊1):385-387.
[13] 徐艳,胡顺波,王基烽,等.一种基于Snake模型的脑部CT图像分割新算法[J].中国医学物理学杂志,2020,37(5):568-573.DOI:10.3969/j.issn.1005-202X.2020.05.007.
[14] 梁华为.直接从双峰直方图确定二值化阈值[J].模式识别与人工智能,2002,15(2):253-256.DOI:10.3969/j.issn.1003-6059.2002.02.022.
[15] 张虎龙.基于二维直方图和模糊熵准则的阈值化方法[J].计算机测量与控制,2017,25(5):236-239.DOI:10.16526/j.cnki.11-4762/tp.2017.05.065.
[16] OTSU N.A threshold selection method from gray-level histograms[J].IEEE Transactions on System Man and Cybernetic,1979,9(1):62-66.DOI:10.1109/TSMC.1979.4310076.
[17] 曾建华,黄时杰.典型图像边缘检测算子的比较与分析[J].河北师范大学学报(自然科学版),2020,44(4):295-301.DOI:10.13763/j.cnki.jhebnu.nse.2020.04.003.
[18] 符尧,王俊峰,高琳,等.基于ASM和GrabCut的DR图像肺实质分割[J].现代计算机(专业版),2015(6):25-30.DOI:10.3969/j.issn.1007-1423.2015.06.006.
[19] 佘广南,陈莹胤,钟丽明,等.基于密集特征匹配的胸片肺野自动分割[J].南方医科大学学报,2016,36(1):61-66.DOI:10.3969/j.issn.1673-4254.2016.01.11.

备注/Memo

备注/Memo:
收稿日期: 2020-12-17
通信作者: 李平(1981-),女,副教授,博士,主要从事非线性系统与智能控制、复杂控制系统的研究.E-mail:pingping_1213@126.com.
基金项目: 国家自然科学基金资助项目(61603144); 福建省自然科学基金资助项目(2018J01095); 福建省高校产学合作科技重大项目(2013H6016); 华侨大学中青年教师科技创新资助计划项目(ZQN-PY509)
更新日期/Last Update: 2022-05-20