一种基于卷积神经网络的胎儿头围自动测量方法

doi:10.12068/j.issn.1005-3026.2023.11.008

东北大学学报（自然科学版） ›› 2023, Vol. 44 ›› Issue (11): 1571-1577.DOI: 10.12068/j.issn.1005-3026.2023.11.008

一种基于卷积神经网络的胎儿头围自动测量方法

杨超然¹，廖珊珊²，陈达¹，康雁^1,3

(1. 东北大学医学与生物信息工程学院，辽宁沈阳110169； 2. 中国医科大学附属盛京医院，辽宁沈阳110801； 3. 深圳技术大学健康与环境工程学院，广东深圳518118)

发布日期:2023-12-05
通讯作者: 杨超然
作者简介:杨超然(1988-)，男，辽宁沈阳人，东北大学博士研究生；康雁(1964-)，男，辽宁沈阳人，东北大学教授，博士生导师.
基金资助:
国家自然科学基金资助项目(62071311)；国家重点研发计划项目(2018YFC1002900).

Automatic Measurement Method for Fetal Head Circumference Based on Convolution Neural Network

YANG Chao-ran¹， LIAO Shan-shan²， CHEN Da¹， KANG Yan^1,3

1. School of Medicine and Biological Information Engineering， Northeastern University， Shenyang 110169， China; 2. Shengjing Hospital of China Medical University， Shenyang 110801， China; 3. School of Health Science Environmental Engineering， Shenzhen Technology University， Shenzhen 518118， China.

Published:2023-12-05
Contact: KANG Yan
About author:-
Supported by:
-

摘要/Abstract

摘要： 在产前超声筛查过程中，为了能够帮助医生在丘脑标准平面上快速、精确地测量胎儿头围，提出一种新颖的双分支卷积神经网络直接分割胎儿颅骨边界，2个分支通过共享层相互促进，有效地提高了颅骨边界的分割精度，特别对局部不清晰或者不连续的边界仍然有着较好的分割效果，具有较高的鲁棒性.本方法的测量过程不需要过多的后处理操作，并且模型属于轻量级网络，便于部署.该方法应用在Grand-Challenge中的HC18数据集及从医院采集的300例数据上，均取得了较好的结果，对比其他主流分割网络如U-Net，Res-U-Net，U-Net++，CE-Net等，所提方法具有更高的分割精度及更小的测量误差.

关键词: 产前超声；头围测量；边界分割；卷积神经网络；轻量级

Abstract: In prenatal ultrasound screening， in order to help doctors measure the fetal head circumference quickly and accurately on the standard plane of the thalamus， a novel two-branch convolution neural network is proposed to directly segment the fetal skull boundary. The two branches promote each other through the shared layer， which can improve the segmentation accuracy of the skull boundary effectively. In particular， the proposed method has good segmentation effects and high robustness for locally unclear or discontinuous boundaries. Furthermore， the measurement process of the proposed method does not require excessive post-processing operations， and the model belongs to a lightweight network， which is easy to deploy. Good results were achieved on the HC18 dataset of Grand-Challenge and 300 cases collected from hospitals. Compared with other mainstream segmentation networks such as U-Net， Res-U-Net， U-Net++， CE-Net， etc.， the proposed method is with higher segmentation accuracy and smaller measurement error.

Key words: prenatal ultrasound; head circumference measurement; boundary segmentation; convolution neural network; lightweight

中图分类号:

TP391

杨超然，廖珊珊，陈达，康雁. 一种基于卷积神经网络的胎儿头围自动测量方法[J]. 东北大学学报（自然科学版）, 2023, 44(11): 1571-1577.

YANG Chao-ran， LIAO Shan-shan， CHEN Da， KANG Yan. Automatic Measurement Method for Fetal Head Circumference Based on Convolution Neural Network[J]. Journal of Northeastern University(Natural Science), 2023, 44(11): 1571-1577.

参考文献

[1]Chudleigh T.The 18⁺⁰-20⁺⁶ weeks fetal anomaly scan national standards［J］.Ultrasound，2010，18(2):92-98.
[2]Sarris I，Ioannou C，Chamberlain P，et al.Intra-and interobserver variability in fetal ultrasound measurements［J］.Ultrasound in Obstetrics & Gynecology，2012，39(3):266-273.
[3]Nadiyah P，Rofiqah N，Firdaus Q，et al.Automatic detection of fetal head using haar cascade and fit ellipse［C］//2019 International Seminar on Intelligent Technology and Its Applications(ISITIA).Piscataway:IEEE，2019:320-324.
[4]van den Heuvel T L A，Dagmar D B，de Korte C L，et al.Automated measurement of fetal head circumference using 2D ultrasound images［J］.PLoS ONE，2018，13(8):0200412.
[5]Zhang L，Dudley N J，Lambrou T，et al.Automatic image quality assessment and measurement of fetal head in two-dimensional ultrasound image［J］.Journal of Medical Imaging，2017，4(2):024001.
[6]Sobhaninia Z，Emami A，Karimi N，et al.Localization of fetal head in ultrasound images by multiscale view and deep neural networks［C］//The 25th International Computer Conference，Computer Society of Iran(CSICC).Piscataway:IEEE，2020:1-5.
[7]Skeika E L，Luz M R，Fernandes B J T，et al.Convolutional neural network to detect and measure fetal skull circumference in ultrasound imaging［J］.IEEE Access，2020，8:191519-191529.
[8]Yang X，Wang X，Wang Y，et al.Hybrid attention for automatic segmentation of whole fetal head in prenatal ultrasound volumes［J］.Computer Methods and Programs in Biomedicine，2020，194:105519.
[9]Grand-Challenge HC18［EB/OL］.(2018-07-13)［2021-10-03］.https://hc18.grand-challenge.org，2018.Z.
[10]Romera E，Alvarez J M，Bergasa L M，et al.ERF-Net:efficient residual factorized convnet for real-time semantic segmentation［J］.IEEE Transactions on Intelligent Transportation Systems，2018，19(1):263-272.
[11]Birant D，Kut A.ST-DBSCAN:an algorithm for clustering spatial-temporal data［J］.Data & Knowledge Engineering，2007，60(1):208-221.
[12]Irene K，Aditya P，Haidi H，et al.Fetal head and abdomen measurement using convolutional neural network，hough transform，and difference of Gaussian revolved along elliptical path(Dogell)algorithm ［EB/OL］.(2019-11-14)［2021-10-08］.https://arxiv.org/abs/1911.06298.
[13]Brabandere B D，Neven D，Gool L V.Semantic instance segmentation for autonomous driving［C］//Computer Vision & Pattern Recognition Workshops.Piscataway:IEEE，2017:7-9.
[14]Ronneberger O，Fischer P，Brox T.U-net:convolutional networks for biomedical image segmentation［C］//Medical Image Computing and Computer-Assisted Intervention(MICCAI).Berlin:Springer International Publishing，2015:234-241.
[15]Xiao X，Shen L，Luo Z M，et al.Weighted Res-UNet for high-quality retina vessel segmentation［C］//2018 9th International Conference on Information Technology in Medicine and Education(ITME).Hangzhou，2018:327-331.
[16]Zhou Z W，Siddiquee M M R，Tajbakhsh N，et al.UNet++:a nested U-net architecture for medical image segmentation［C］//The 4th Deep Learning in Medical Image Analysis(DLMIA)Workshop.Berlin:Springer International Publishing，2018:3-11.
[17]Gu Z W，Cheng J，Fu H Z，et al.CE-Net:context encoder network for 2D medical image segmentation［J］.IEEE Transactions on Medical Imaging，2019，38(10):2281-2292.
[18]Badrinarayanan V，Kendall A，Cipolla R.Segnet:a deep convolutional encoder-decoder architecture for image segmentation［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence，2017，39(12):2481-2495.
[19]Qiao D H，Zulkernine F.Dilated squeeze-and-excitation U-net for fetal ultrasound image segmentation［C］//2020 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology(CIBCB).Piscataway:IEEE，2020:1-7.

一种基于卷积神经网络的胎儿头围自动测量方法

Automatic Measurement Method for Fetal Head Circumference Based on Convolution Neural Network

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	李佳洋, 唐加福, 郭恩铭. 基于动态情绪驱动的人员应急疏散模型[J]. 东北大学学报（自然科学版）, 2021, 42(11): 1656-1662.
[2]	杨丹，刘国如，任梦成，裴宏杨. 多尺度卷积核U-Net模型的视网膜血管分割方法[J]. 东北大学学报（自然科学版）, 2021, 42(1): 7-14.
[3]	韩东红，张宏亮，朱帅伟，齐孝龙. 面向新浪微博的情感社区检测算法[J]. 东北大学学报（自然科学版）, 2021, 42(1): 21-31.
[4]	原培新，陈鼎夫. 双能X射线高动态范围安检图像压缩算法[J]. 东北大学学报（自然科学版）, 2021, 42(1): 96-101.
[5]	魏颖, 徐楚翘, 刁兆富, 李伯群. 基于生成对抗网络的多目标行人跟踪算法[J]. 东北大学学报（自然科学版）, 2020, 41(12): 1673-1680.
[6]	李稷，徐安军. 炼钢车间多天车动态调度仿真方案[J]. 东北大学学报:自然科学版, 2020, 41(12): 1699-1707.
[7]	李占山，吕艾娜. 基于新冗余度的特征选择方法[J]. 东北大学学报:自然科学版, 2020, 41(11): 1550-1556.
[8]	王鑫，王翠荣，王聪，苑迎. 双通道多感知卷积神经网络图像超分辨率重建[J]. 东北大学学报:自然科学版, 2020, 41(11): 1564-1570.
[9]	张春雷，戴丽，刘宇，李鹤. 基于三点法和ICP算法的手术导航系统患者配准[J]. 东北大学学报:自然科学版, 2020, 41(11): 1584-1590.
[10]	陈剑，何涛，闻英友，马林涛. 基于BERT模型的司法文书实体识别方法[J]. 东北大学学报:自然科学版, 2020, 41(10): 1382-1387.
[11]	张田，田勇，王子，王昭东. 基于清晰度评价的自适应阈值图像分割法[J]. 东北大学学报:自然科学版, 2020, 41(9): 1231-1238.
[12]	张丽杰，刘建昌，谭树彬. 复杂建筑火灾中的人员疏散路径多目标规划[J]. 东北大学学报:自然科学版, 2020, 41(6): 761-766.
[13]	赵海，周冰玲，朱宏博，窦圣昶. 基于连续最大流的三维肺实质快速分割算法[J]. 东北大学学报:自然科学版, 2020, 41(4): 470-474.
[14]	徐久强，张金鹏，贾玉其，邵建新. 基于集成学习的束支传导阻滞识别方法[J]. 东北大学学报:自然科学版, 2020, 41(3): 321-326.
[15]	孟琭，钟健平，李楠. 基于GAN的医学图像仿真数据集生成算法[J]. 东北大学学报:自然科学版, 2020, 41(3): 332-336.