期刊文献+

基于深度卷积神经网络的CT图像层间插值方法的初步研究 被引量:1

Primary Study on Inter-Layer Interpolation Method of CT Image Based on Deep Convolutional Neural Network
下载PDF
导出
摘要 目的:研究一种基于CT图像层间插值的方法,用于放射治疗过程中的患者摆位验证,从而提高放疗精度。方法:采用一种基于3D卷积和膨胀卷积神经网络(3D CNN-DCNN)算法,利用相邻图像层之间的关联信息重建中间层图像。采用U-Net网络架构,通过编码部分的卷积层、膨胀卷积层、池化层和解码部分的上采样层、卷积层、膨胀卷积层,对CT进行端到端的学习。采集20例患者图像数据,采用留一交叉验证的方法训练验证模型,分别对神经网络和线性插值的预测CT与原始薄层CT进行对照比较。结果:3D CNN-DCNN的平均绝对误差(MAE)为34 HU,远小于线性插值的55 HU。除此之外,骨骼的Dice相似系数(DSC)为0.95,大于线性插值方法的0.89。结论:与传统线性插值方法相比,3D CNN-DCNN算法可以更准确的重建薄层CT,明显改善了插值伪影、图像失真和锯齿状现象。 Purpose:To study a method based on interslice interpolation of CT images for the verification of patient setup during radiotherapy,and to improve the accuracy of radiotherapy.Methods:An algorithm based on 3D convolution and dilated convolutional neural network(3D CNN-DCNN)was used to reconstruct the intermediate slice images of CT by utilizing the correlation information between adjacent image slices.By using the U-Net network structure,end-to-end learning was done on CT images by convolutional layer,dilated convolutional layer,pooling layer in the encoding portion and up sampling layer,convolutional layer,and dilated convolutional layer in the decoding portion.The neural network was trained and tested with the image data of 20 patients using the method of leave-one-out cross validation.The predicted CT images from the neural network and linear interpolation were compared with the original thin-slice CT respectively.Results:The mean absolute error(MAE)of 3D CNN-DCNN is 34 HU,which was much smaller than that of linear interpolation(55 HU).The Dice similarity coefficient(DSC)of the bones was 0.95,which was larger than that of the linear interpolation method(0.89).Conclusion:Compared with linear interpolation method,the 3D CNN-DCNN algorithm can be used to reconstruct thin-slice CT more accurately,and improves the interpolation artifacts,image distortion and jaggedness.
作者 菅影超 马善达 王伟 JIAN Yingchao;MA Shanda;WANG Wei(R&D Department,Bejing Rayer Shiwei Medical Research Co Ltd.;R&D Department,Jiangou Rayer Medica!Technology Co.,Ltd.;Department of Radiotherapy,Tianjin Medical University Cancer Hospital)
出处 《中国医学计算机成像杂志》 CSCD 北大核心 2022年第6期669-675,共7页 Chinese Computed Medical Imaging
关键词 摆位验证 膨胀卷积神经网络 U-Net 留一交叉验证 平均绝对误差 Dice相似系数 Setup verification Dilated convolutional neural network U-Net Leave-one-out cross validation Mean absolute error Dice similarity coefficient
  • 相关文献

参考文献4

二级参考文献22

  • 1章毓晋.图像工程(上册)[M].北京:清华大学出版社,1999.201-204.
  • 2Kim J.Effects of X-ray and CT image enhancements on the robustness and accuracy of a rigid 3D/2D image registration[J].Med Phys,2005,32(4):866-873.
  • 3Eilers PH.Enhancing scatterplots with smoothed densities[J].Bioinformatics,2004,20(5):623-628.
  • 4Moler C.Using matlab graphis[M].6th ed.The MathWorks Inc,2002:102-198.
  • 5Jensen JA,Holm O,Jensen LJ,et al.Ultrasound research scanner for real-time synthetic aperture data acquisition[J].IEEE Trans Ultrason Ferroel Freq Control,2005,52(5):881-891.
  • 6Stark JA.Adaptive image contrast enhancement using generalizations of histogramequalization[J].IEEE Trans Image Proc,2000,9(5):889-996.
  • 7Wang L.Enhancement of medical ultrasonic image based on gray-level histogram equalization[J].J Sichuan Univ,2002,34(1):105-108.
  • 8Saffor A,bin Ramli AR,Ng KH.Wavelet-based compression of medical images:filter-bank selection and evaluation[J].Australas Phys Eng Sci Med,2003,26(2):39-44.
  • 9余凯,贾磊,陈雨强,徐伟.深度学习的昨天、今天和明天[J].计算机研究与发展,2013,50(9):1799-1804. 被引量:610
  • 10刘建伟,刘媛,罗雄麟.玻尔兹曼机研究进展[J].计算机研究与发展,2014,51(1):1-16. 被引量:71

共引文献1768

同被引文献5

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部