A generative flow-based model for volumetric data augmentation in 3D deep learning for computed tomographic colonography.
A generative flow-based model for volumetric data augmentation in 3D deep learning for computed tomographic colonography.
复制标题
DOI：
10.1007/s11548-020-02275-z
复制
发表时间：
2021-01
影响因子：
3
通讯作者：
Yoshida H
中科院分区：
文献类型：
--
作者：
Uemura T;Näppi JJ;Ryu Y;Watari C;Kamiya T;Yoshida H
深度学习可用于提高计算机辅助检测（CADe）在各种医学成像任务中的性能。然而，在计算机断层扫描（CT）结肠镜检查中，性能受到相对较小的尺寸和可用训练数据集的多样性的限制。我们在这项研究中的目的是开发和评估一种基于流量的生成模型，用于进行结肠直肠息肉的3D数据增强，以有效训练CADe中用于CT结肠镜检查的深度学习。我们基于基于流的生成模型（3D Glow）开发了一个3D卷积神经网络（3D CNN），用于生成合成感兴趣体积（voi），其特征与其训练数据集的voi相似。利用我们收集的临床CT结肠镜病例，对3D Glow进行训练，生成息肉的合成voi。评估是通过使用三名观察者的人类观察者研究和使用基于cad的息肉分类研究和3D DenseNet进行的。三名观察者的受试者工作特征分析曲线下面积值在区分真实息肉和合成息肉方面无统计学差异。当使用3D Glow进行数据增强训练时，3D DenseNet的息肉分类性能显著高于使用其他增强方法进行训练时的性能。3D发光生成的合成息肉在视觉上与真正的结肠直肠息肉难以区分。将其应用于数据增强，可以大大提高三维cnn在CT结肠镜CADe中的性能。因此，3D Glow是一种很有前途的方法，可以提高CADe中CT结肠镜的深度学习性能。
Deep learning can be used for improving the performance of computer-aided detection (CADe) in various medical imaging tasks. However, in computed tomographic (CT) colonography, the performance is limited by the relatively small size and the variety of the available training datasets. Our purpose in this study was to develop and evaluate a flow-based generative model for performing 3D data augmentation of colorectal polyps for effective training of deep learning in CADe for CT colonography. We developed a 3D-convolutional neural network (3D CNN) based on a flow-based generative model (3D Glow) for generating synthetic volumes of interest (VOIs) that has characteristics similar to those of the VOIs of its training dataset. The 3D Glow was trained to generate synthetic VOIs of polyps by use of our clinical CT colonography case collection. The evaluation was performed by use of a human observer study with three observers and by use of a CADe-based polyp classification study with a 3D DenseNet. The area-under-the-curve values of the receiver operating characteristic analysis of the three observers were not statistically significantly different in distinguishing between real polyps and synthetic polyps. When trained with data augmentation by 3D Glow, the 3D DenseNet yielded a statistically significantly higher polyp classification performance than when it was trained with alternative augmentation methods. The 3D Glow-generated synthetic polyps are visually indistinguishable from real colorectal polyps. Their application to data augmentation can substantially improve the performance of 3D CNNs in CADe for CT colonography. Thus, 3D Glow is a promising method for improving the performance of deep learning in CADe for CT colonography.
登录
查看更多内容
影响因子：
158.5
作者：
Pickhardt, PJ;Choi, JR;Schindler, WR
通讯作者：
Schindler, WR
影响因子：
4.8
作者：
Näppi, J;Yoshida, H
通讯作者：
Yoshida, H
影响因子：
8.1
作者：
Shorten, Connor;Khoshgoftaar, Taghi M.
通讯作者：
Khoshgoftaar, Taghi M.
DOI：
10.1056/nejmoa0800996
发表时间：
2008-09-18
期刊：
The New England journal of medicine
影响因子：
--
作者：
Johnson CD;Chen MH;Toledano AY;Heiken JP;Dachman A;Kuo MD;Menias CO;Siewert B;Cheema JI;Obregon RG;Fidler JL;Zimmerman P;Horton KM;Coakley K;Iyer RB;Hara AK;Halvorsen RA Jr;Casola G;Yee J;Herman BA;Burgart LJ;Limburg PJ
通讯作者：
Limburg PJ
影响因子：
4.8
作者：
Nagata, Koichi;Okawa, Tomohiko;Yoshida, Hiroyuki
通讯作者：
Yoshida, Hiroyuki