Fig 1.
Schematic illustration of the modified VGG-16.
Note: Except softmax layer, activation function is not shown.
Table 1.
Summary of patient demographics.
Fig 2.
Representative CT images of lung nodules.
(A) benign nodule, (B) primary lung cancer and (C) metastatic lung cancer.
Fig 3.
Three CT images obtained from three orthogonal planes used for input to 2D-DCNN.
Fig 2(B) is identical to Fig 3(A). (A) axial image, (B) coronal image and (C) sagittal image. Abbreviations: DCNN, deep convolutional neural network.
Table 2.
Optimal hyperparameters and classification results for CADx by DCNN with and without transfer learning.
Fig 4.
Representative results of loss and accuracy during DCNN training with transfer learning.
Abbreviations: DCNN, deep convolutional neural network.
Fig 5.
Representative results of loss and accuracy during DCNN training without transfer learning.
Abbreviations: DCNN, deep convolutional neural network.
Table 3.
Representative result of confusion matrix between true labels and predicted labels by DCNN with transfer learning.
Table 4.
Representative result of confusion matrix between true labels and predicted labels by DCNN without transfer learning.
Table 5.
Result of averaged confusion matrix between true labels and predicted labels by DCNN with transfer learning.