[1]林家庆,韩娟,袁直敏,等.多任务深度卷积网络的CT图像方向校正[J].华侨大学学报(自然科学版),2020,41(3):366-373.[doi:10.11830/ISSN.1000-5013.201911039]
 LIN Jiaqing,HAN Juan,YUAN Zhimin,et al.Orientation Correction for CT Images via Multitask Deep Convolutional Network[J].Journal of Huaqiao University(Natural Science),2020,41(3):366-373.[doi:10.11830/ISSN.1000-5013.201911039]
点击复制

多任务深度卷积网络的CT图像方向校正()
分享到:

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

卷:
第41卷
期数:
2020年第3期
页码:
366-373
栏目:
出版日期:
2020-05-20

文章信息/Info

Title:
Orientation Correction for CT Images via Multitask Deep Convolutional Network
文章编号:
1000-5013(2020)03-0366-08
作者:
林家庆 韩娟 袁直敏 彭佳林
华侨大学 计算机科学与技术学院, 福建 厦门 361021
Author(s):
LIN Jiaqing HAN Juan YUAN Zhimin PENG Jialin
College of Computer Science and Technology, Huaqiao University, Xiamen 361021, China
关键词:
CT图像 方向校正 深度卷积网络 多任务回归网络
Keywords:
CT image orientation correction deep convolution network multitask regression network
分类号:
TP183;R814.42
DOI:
10.11830/ISSN.1000-5013.201911039
文献标志码:
A
摘要:
针对医学电子计算机断层扫描(CT)图像方向校正问题,提出一种并行卷积回归(PCRN)多任务深度学习网络.通过侧旋角度正回归和翻转概率逻辑回归,求得校正参数来精准地校正图像.进一步,针对医学图像训练样本稀缺的情况,提出一种串行回归(SCRN)的深度学习架构,弥补并行卷积回归网络在小样本情况下校正精度不足的问题.实验结果表明:在样本充分,并行卷积回归网络和样本稀缺情况下,串行卷积回归网络对小角度偏转、大角度偏转和翻转的腹部CT图像校正结果都优于传统的配准方法.
Abstract:
To address the problem of orientation correction of CT image, a parallel convolution regression network(PCRN)with multitask deep learning is proposed. Orientation parameters are learned by the positive regression of the lateral rotation angle and the reversal probability logistic regression to accurate calibration images. Furthermore, in view of the lack in training samples for medical images, a deep learning frame named serial convolution regression network(SCRN)is introduced, which makes up for the inadequate correction accuracy of parallel convolutional regression network in the case of small samples. The experimental results show that the PCRN method with sufficient samples and the SCRN method with scarce samples are superior to the traditional registration methods in correcting CT images with small, large angles and flipped situation.

参考文献/References:

[1] TRIVEDI D N,SHAH N D,KOTHARI A M,et al.DICOM</sup>○R medical image standard[M]//Déntal Image Processing for Human Identification.Berlin: Springer, Cham,2019:41-49.DOI: 10.1007/978-3-319-99471-0_4.
[2] SONG Guoli,HAN Jianda,ZHAO Yiwen,et al.A review on medical image registration as an optimization problem[J].Current Medical Imaging Reviews,2017,13(3):274-283.DOI:10.2174/1573405612666160920123955.
[3] WEI Weimin Wei,WANG Shuozhong,ZHANG Xinpeng,et al.Estimation of image rotation angle using interpolation-related spectral signatures with application to blind detection of image forgery[J].IEEE Transactions on Information Forensics and Security,2010,5(3):507-517.DOI:10.1109/tifs.2010.2051254.
[4] SOLANKI K,MADHOW U,MANJUNATH B S,et al.Estimating and undoing rotation for print-scan resilient data hiding[C]//International Conference on Image Processing.Singapore:IEEE Press,2004:39-42.DOI:10.1109/ICIP.2004.1418684.
[5] FEFILATYEV S, SMARODZINAVA V, HALL L O,et al.Horizon detection using machine learning techniques[C]//International Conference on Machine Learning and Applications.Orlando:IEEE Press,2006:17-21.DOI:10.1109/ICMLA.2006.25.
[6] VAILAYA A,ZHANG Hongjiang,YANG Changjiang,et al.Automatic image orientation detection[J].IEEE Transactions on Image Processing,2002,11(7):746-755.DOI:10.1109/tip.2002.801590.
[7] ESTEVA A,ROBICQUET A,RAMSUNDAR B,et al.A guide to deep learning in healthcare[J].Nature Medicine,2019,25(1):24-29.DOI:10.1038/s41591-018-0316-z.
[8] PRINCE M,ALSUHIBANY S A,SIDDIQI N A.A step towards the optimal estimation of image orientation[J].IEEE Access,2019,7:185750-185759.
[9] OSADCHY M,CUN Y L,MILLER M L.Synergistic face detection and pose estimation with energy-based models[J].Journal of Machine Learning Research,2007,8:1197-1215.
[10] BALTRUSCHAT I M,SAALBACH A,HEINRICH M P,et al.Orientation regression in hand radiographs:A transfer learning approach[C]//Medical Imaging 2018:Image Processing.Houston:SPIE,2018.DOI:10.1117/12.2291620.
[11] SAFONOV I V,KURILIN I V,RYCHAGOV M N,et al.Content-based image orientation recognition[M]//Adaptive Image Processing Algorithms for Printing.Singapore:Springer,2018:269-277.DOI:10.1007/978-981-10-6931-4_12.
[12] MORRA L,FAMOURI S,KARAKUS H C,et al.Automatic detection of canonical image orientation by convolutional neural networks[C]//IEEE 23rd International Symposium on Consumer Technologies(ISCT).Ancona:IEEE Press,2019:118-123.
[13] FISCHER P,DOSOVITSKIY A,BROX T.Image orientation estimation with convolutional networks[C]//German Conference on Pattern Recognition.Switzerland:Springer,2015:368-378.DOI:10.1007/978-3-319-24947-6_30.
[14] de VOS B D,BERENDSEN F F,VIERGEVER M A,et al.A deep learning framework for unsupervised affine and deformable image registration[J].Medical Image Analysis,2019,52:128-143.DOI:10.1016/j.media.2018.11.010.
[15] EPPENHOF K A J,LAFARGE M W,VETA M,et al.Progressively trained convolutional neural networks for deformable image registration[EB/OL].[2019-11-15] .https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8902170.DOI:10.1109/TMI.2019.2953788.
[16] HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al.Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas:IEEE Press,2016:770-778.DOI:10.1109/CVPR.2016.90.
[17] ZILLY J G,SRIVASTAVA R K,KOUTNÍK J,et al.Recurrent highway networks[C]//Proceedings of the 34 th International Conference on Machine Learning.Sydney:IEEE Press,2017:4189-4198.
[18] HUANG Gao,LIU Zhuang,PLEISS G,et al.Convolutional networks with dense connectivity[EB/OL].[2019-05-23] .https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8721151.DOI:10.1109/TPAMI.2019.2918284.
[19] LIN C H,LUCEY S.Inverse compositional spatial transformer networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Honolulu:IEEE Press,2017:2568-2576.DOI:10.1109/CVPR.2017.242.
[20] KINGMA D P,BA J.Adam:A method for stochastic optimization[EB/OL].(2014-12-22)[2017-01-30] .https://arxiv.org/abs/1412.6980.

备注/Memo

备注/Memo:
收稿日期: 2019-11-18
通信作者: 彭佳林(1985-),男,副教授,博士,主要从事机器学习及应用、图像处理和神经影像分析的研究.E-mail:2004pjl@163.com.
基金项目: 国家自然科学基金资助项目(11771160)
更新日期/Last Update: 2020-05-20