[1]何奕松,蒋家良,余行,等.影像分割中Dice系数和Hausdorff距离的比较[J].中国医学物理学杂志,2019,36(11):1307-1311.[doi:DOI:10.3969/j.issn.1005-202X.2019.11.012]
　HE Yisong,JIANG Jialiang,YU Hang,et al.Comparison of Dice similarity coefficient and Hausdorff distance in image segmentation[J].Chinese Journal of Medical Physics,2019,36(11):1307-1311.[doi:DOI:10.3969/j.issn.1005-202X.2019.11.012]
点击复制

影像分割中Dice系数和Hausdorff距离的比较()
分享到：

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

卷:
36卷
期数:
2019年第11期
页码:
1307-1311
栏目:
医学影像物理
出版日期:
2019-11-25

文章信息/Info

Title:
Comparison of Dice similarity coefficient and Hausdorff distance in image segmentation
文章编号:
1005-202X（2019）11-1307-05
作者:
何奕松; 蒋家良; 余行; 傅玉川
四川大学华西医院放疗科， 四川 成都 610041
Author(s):
HE Yisong;  JIANG Jialiang;  YU Hang;  FU Yuchuan
Department of Radiotherapy, West China Hospital, Sichuan University, Chengdu 610041, China
关键词:
精确放疗; 靶区勾画; 轮廓相似性; Dice系数; Hausdorff距离
Keywords:
precision radiotherapy;  segmentation of target area;  contour similarity;  Dice similarity coefficient;  Hausdorff distance
分类号:
R318
DOI:
DOI:10.3969/j.issn.1005-202X.2019.11.012
文献标志码:
A
摘要:
比较影像分割技术中Dice系数和Hausdorff距离的评价效果，分析这两种相似性系数之间的联系，并设计相应的类型加以进一步论述。方法：采用Dice系数和Hausdorff距离对两种轮廓相似度进行评估。设计18个（9对）从临床靶区中抽象出的轮廓，计算并比较这两类相似性系数。结果：根据比较结果，总结出3种不同类型情况：（1）图像符合度好；（2）图像整体符合度好，但存在一小部分符合度差；（3）图像轮廓符合度差。结论：仅靠Dice系数作为唯一评判标准是不完善的，应在此基础上增加Hausdorff距离的筛选，这样才能最大限度地说明轮廓之间的符合程度。
Abstract:
Objective To compare the evaluation effects of Dice similarity coefficient and Hausdorff distance in image segmentation; analyze the relationships between two similarity coefficients; and design the corresponding contours for further discussion. Methods Dice similarity coefficient and Hausdorff distance were used to evaluate contour similarity. Moreover, 18 (9 pairs) virtual contours abstracted from clinical target areas were designed for calculating and comparing the two similarity coefficients, namely Dice similarity coefficient and Hausdorff distance. Results According to the comparison results, 3 different types of cases were summarized. (1) The image conformity was preferable; (2) the overall conformity of the images was good, but there was a small proportion of poor conformity; (3) the conformity of image contour was poor. Conclusion Using Dice similarity coefficient as the only criterion is imperfect, and Dice similarity coefficient should be combined with Hausdorff distance to reflect contour conformity to the greatest extent.

相似文献/References:

[1]彭莹莹,张书旭,谭剑明,等.基于PCNN的PET/CT图像分割在肺癌靶区勾画中的应用[J].中国医学物理学杂志,2014,31(04):5022.[doi:10.3969/j.issn.1005-202X.2014.04.011]
[2]刘国才,胡泽田,朱苏雨,等.头颈部肿瘤PET与MRI融合放疗靶区自适应区域生长勾画[J].中国医学物理学杂志,2016,33(3):222.[doi:10.3969/j.issn.1005-202X.2016.03.002]
　[J].Chinese Journal of Medical Physics,2016,33(11):222.[doi:10.3969/j.issn.1005-202X.2016.03.002]
[3]陈绍芳. 锥形束CT引导乳腺癌精确放疗的研究进展[J].中国医学物理学杂志,2019,36(2):163.[doi:DOI:10.3969/j.issn.1005-202X.2019.02.008]
　CHEN Shaofang. Research progress on cone-beam CT image-guided precise radiotherapy of breast cancer[J].Chinese Journal of Medical Physics,2019,36(11):163.[doi:DOI:10.3969/j.issn.1005-202X.2019.02.008]
[4]蒋家良,罗勇,何奕松,等.特征区域再聚焦提升全卷积神经网络勾画较小靶区准确度[J].中国医学物理学杂志,2020,37(1):75.[doi:DOI:10.3969/j.issn.1005-202X.2020.01.015]
　JIANG Jialiang,LUO Yong,HE Yisong,et al.Feature area refocusing for improving the accuracy of small target area segmentations by fully convolutional networks[J].Chinese Journal of Medical Physics,2020,37(11):75.[doi:DOI:10.3969/j.issn.1005-202X.2020.01.015]
[5]成肇杰,韦婷,梁卫学,等.CARE Dose 4D在鼻咽癌放疗定位中的应用[J].中国医学物理学杂志,2021,38(11):1349.[doi:DOI:10.3969/j.issn.1005-202X.2021.11.006]
　CHENG Zhaojie,WEI Ting,LIANG Weixue,et al.Application of CARE Dose 4D in radiotherapy localization of nasopharyngeal carcinoma[J].Chinese Journal of Medical Physics,2021,38(11):1349.[doi:DOI:10.3969/j.issn.1005-202X.2021.11.006]
[6]申海洋,彭祥炜,王兴,等.一种基于肝分段和肝癌轮廓融合的放疗靶区数据库建立方法[J].中国医学物理学杂志,2022,39(10):1250.[doi:DOI:10.3969/j.issn.1005-202X.2022.10.012]
　SHEN Haiyang,PENG Xiangwei,et al.A method for establishing radiotherapy target area database based on the fusion of liver segment and liver cancer contour[J].Chinese Journal of Medical Physics,2022,39(11):1250.[doi:DOI:10.3969/j.issn.1005-202X.2022.10.012]
[7]郭翌,蓝林臻,刘清泉,等.两种自动勾画方法对T3分期鼻咽癌靶区勾画结果对比分析[J].中国医学物理学杂志,2023,40(3):291.[doi:DOI:10.3969/j.issn.1005-202X.2023.03.006]
　GUO Yi,,et al.Comparison of target segmentation results of two auto-segmentation methods for T3 nasopharyngeal carcinoma[J].Chinese Journal of Medical Physics,2023,40(11):291.[doi:DOI:10.3969/j.issn.1005-202X.2023.03.006]

备注/Memo

备注/Memo:
【收稿日期】2019-06-10 【作者简介】何奕松，在读硕士，研究方向：医学物理，E-mail:627412679- @qq.com 【通信作者】傅玉川，博士，主任技师，研究方向：医学物理，E-mail: ychfu@hotmail.com
更新日期/Last Update: 2019-11-28