Bermejo-Peláez David, San José Estépar Raúl, Ledesma-Carbayo M J
Biomedical Image Technologies, Universidad Politécnica de Madrid & CIBER-BBN, Madrid, Spain.
Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA.
Proc IEEE Int Symp Biomed Imaging. 2018 Apr;2018:519-522. doi: 10.1109/isbi.2018.8363629. Epub 2018 May 24.
In this article we propose and validate a fully automatic tool for emphysema classification in Computed Tomography (CT) images. We hypothesize that a relatively simple Convolutional Neural Network (CNN) architecture can learn even better discriminative features from the input data compared with more complex and deeper architectures. The proposed architecture is comprised of only 4 convolutional and 3 pooling layers, where the input corresponds to a 2.5D multiview representation of the pulmonary segment tissue to classify, corresponding to axial, sagittal and coronal views. The proposed architecture is compared to similar 2D CNN and 3D CNN, and to more complex architectures which involve a larger number of parameters (up to six times larger). This method has been evaluated in 1553 tissue samples, and achieves an overall sensitivity of 81.78 % and a specificity of 97.34%, and results show that the proposed method outperforms deeper state-of-the-art architectures particularly designed for lung pattern classification. The method shows satisfactory results in full-lung classification.
在本文中，我们提出并验证了一种用于计算机断层扫描（CT）图像中肺气肿分类的全自动工具。我们假设，与更复杂、更深的架构相比，相对简单的卷积神经网络（CNN）架构能够从输入数据中学习到更好的判别特征。所提出的架构仅由4个卷积层和3个池化层组成，其中输入对应于要分类的肺段组织的2.5D多视图表示，分别对应轴向、矢状和冠状视图。将所提出的架构与类似的2D CNN和3D CNN以及涉及更多参数（多达六倍）的更复杂架构进行了比较。该方法已在1553个组织样本中进行了评估，总体灵敏度达到81.78%，特异性达到97.34%，结果表明所提出的方法优于专门为肺模式分类设计的更深层次的先进架构。该方法在全肺分类中显示出令人满意的结果。
