|本期目录/Table of Contents|

[1]周成礼,罗娜,倪东,等.基于最大熵的超声图像中胎儿股骨分割方法的研究*[J].生物医学工程研究,2017,03:197-200.
 ZHOU Chengli,LUO Na,NI Dong,et al.Automatic Measurement of Fetal Femur Length in Ultrasound Image[J].Journal of Biomedical Engineering Research,2017,03:197-200.
点击复制

基于最大熵的超声图像中胎儿股骨分割方法的研究*(PDF)

《生物医学工程研究》[ISSN:1006-6977/CN:61-1281/TN]

期数:
2017年03期
页码:
197-200
栏目:
出版日期:
2017-09-25

文章信息/Info

Title:
Automatic Measurement of Fetal Femur Length in Ultrasound Image
文章编号:
1672-6278 (2017)03-0197-04
作者:
周成礼1 罗娜2倪东2邓云2△
1.深圳市妇幼保健院超声科,深圳 518000;2. 医学超声关键技术国家地方联合工程实验室, 广东省生物医学信息检测与超声成像重点实验室,深圳大学医学部生物医学工程学院,深圳 518060
Author(s):
ZHOU Chengli1LUO Na2NI Dong2DENG Yun2
1. Shenzhen Maternity&Child Healthcare Hospital, Shenzhen 518000; 2. National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, Guangdong Key Laboratory for Biomedical Measurements and Ultrasound Imaging, School of Biomedical Engineering, Shenzhen University, Shenzhen 518060,China
关键词:
超声图像处理股骨长测量自动测量最大熵分割
Keywords:
Ultrasound Image processing Femur length measurementAutomatic measurementMaximum entropy segmentation
分类号:
R318
DOI:
10.19529/j.cnki.1672-6278.2017.03.02
文献标识码:
A
摘要:
提出一种基于最大熵分割的胎儿股骨自动测量方法。首先,用中值滤波器对原始图像进行去噪,并用最大熵分割方法对去噪后的图像进行分割,得到股骨候选区域;其次,利用股骨区域位置、形状等特征信息对股骨候选区域进行筛选,得到最终的股骨区域;最后,通过股骨区域的外接矩形斜边长,计算股骨长度;与医生手动测量结果对比,70幅超声图像的自动测量结果平均相对误差为1.42±4.48 mm,实验结果验证了本方法的可行性。
Abstract:
To propose a novel automatic method to measure the fetal femur length. First, the candidate regions containing the femur were detected in the ultrasound image using maximum entropy segmentation and gray information. Then the femur region was localized based on both the shape and position of the candidate regions. Finally, the femur end points were detected by hypotenuse of enclosing rectangle of femur and the femur length was measured. Comparing with the results measured by doctors, the average error of 70 ultrasound femur images is 1.42±4.48 mm, which indicates our method can accurately measure the femur length.

参考文献/References

备注/Memo

备注/Memo:
(收稿日期:2017-05-10) 国家自然科学基金资助项目 (6157010571)△通信作者Email:dengyun@szu.edu.cn
更新日期/Last Update: 2017-09-21