Automatic delineation of ribs and clavicles in chest radiographs using fully convolutional DenseNets

Automatic delineation of ribs and clavicles in chest radiographs using fully convolutional DenseNets
复制标题

使用全卷积 DenseNet 自动描绘胸部 X 光照片中的肋骨和锁骨

DOI:
10.1016/j.cmpb.2019.105014
复制
发表时间:
2019-10
影响因子:
6.1
通讯作者:
Yang Wei
Yang Wei
中科院分区:
工程技术2区
文献类型:
--
作者:
Liu Yunbi;Zhang Xiao;Cai Guangwei;Chen Yingyin;Yun Zhaoqiang;Feng Qianjin;Yang Wei

文献摘要

参考文献

相似文献

在胸部X光片(CXR)中,所有的骨骼和软组织都相互重叠,这给放射科医生阅读和解释CXR带来了问题。对肋骨和锁骨的描绘有助于从胸片中抑制它们,从而可以减少它们对胸片分析的影响。然而,通过没有深度学习模型的方法自动描绘肋骨和锁骨是困难的。此外,很少有没有深度学习模型的方法可以有效地描绘前肋骨,因为它们在后前(PA)CXRs.MethodsIn这项工作中的模糊肋骨边缘,我们提出了一种有效的深度学习方法,用于自动描绘后肋骨,前肋骨和锁骨,使用完全卷积的DenseNet(FC-DenseNet)作为像素分类器。我们考虑一个像素加权损失函数,以减轻手动划定鲁棒prediction.ResultsWe进行比较分析与其他两个完全卷积网络的边缘检测和国家的最先进的方法没有深度学习模型的不确定性问题。所提出的方法显着优于这些方法的定量评价指标和视觉感知。在测试数据集上,该方法的平均召回率为0.773 ± 0.030,平均精确率为0.861 ± 0.043,平均边界距离为0.855 ± 0.642像素,平均F值为0.814 ± 0.023。该方法在JSRT和NIH胸部X射线数据集上也表现良好,表明其在多个数据库中的通用性。此外,使用我们的勾画系统还产生了抑制CXR骨骼成分的初步结果。结论所提出的方法可以自动勾画CXR中的肋骨和锁骨,并产生准确的边缘图。
Background and ObjectiveIn chest radiographs (CXRs), all bones and soft tissues are overlapping with each other, which raises issues for radiologists to read and interpret CXRs. Delineating the ribs and clavicles is helpful for suppressing them from chest radiographs so that their effects can be reduced for chest radiography analysis. However, delineating ribs and clavicles automatically is difficult by methods without deep learning models. Moreover, few of methods without deep learning models can delineate the anterior ribs effectively due to their faint rib edges in the posterior-anterior (PA) CXRs.MethodsIn this work, we present an effective deep learning method for delineating posterior ribs, anterior ribs and clavicles automatically using a fully convolutional DenseNet (FC-DenseNet) as pixel classifier. We consider a pixel-weighted loss function to mitigate the uncertainty issue during manually delineating for robust prediction.ResultsWe conduct a comparative analysis with two other fully convolutional networks for edge detection and the state-of-the-art method without deep learning models. The proposed method significantly outperforms these methods in terms of quantitative evaluation metrics and visual perception. The average recall, precision and F-measure are 0.773 ± 0.030, 0.861 ± 0.043 and 0.814 ± 0.023 respectively, and the mean boundary distance (MBD) is 0.855 ± 0.642 pixels of the proposed method on the test dataset. The proposed method also performs well on JSRT and NIH Chest X-ray datasets, indicating its generalizability across multiple databases. Besides, a preliminary result of suppressing the bone components of CXRs has been produced by using our delineating system.ConclusionsThe proposed method can automatically delineate ribs and clavicles in CXRs and produce accurate edge maps.
DOI: 10.1007/978-3-0348-5767-3
发表时间: 1975-09
期刊: --
影响因子: --
作者:
H. Wechsler;J. Sklansky
通讯作者: H. Wechsler;J. Sklansky
DOI: 10.1007/978-3-540-30126-4_14
发表时间: 2004-09
期刊: --
影响因子: --
作者:
R. Moreira;A. Mendonça;A. Campilho
通讯作者: R. Moreira;A. Mendonça;A. Campilho
DOI: 10.1007/s11263-017-1004-z
发表时间: 2017-12-01
影响因子: 19.5
作者:
Xie, Saining;Tu, Zhuowen
通讯作者: Tu, Zhuowen
DOI: 10.1016/0146-664x(82)90016-8
发表时间: 1977-08
期刊: --
影响因子: --
作者:
C. Brace;J. Kulick;T. Challis
通讯作者: C. Brace;J. Kulick;T. Challis
DOI: 10.1007/978-3-642-15549-9_1
发表时间: 2010-09
影响因子: 23.6
作者:
Kaiming He;Jian Sun-;Xiaoou Tang
通讯作者: Kaiming He;Jian Sun-;Xiaoou Tang