Skip to main content
. 2018 Dec;10(12):6624–6635. doi: 10.21037/jtd.2018.11.03

Figure 3.

Figure 3

The AUC of each model to predict EGFR mutations. (A) The AUC of MRadiomics and MRadiomics+Clinical is 0.740 (95% CI, 0.670–0.811) and 0.758 (95% CI, 0.690–0.825); (B) the AUC of MMCNNs and MMCNNs+Clinical is 0.810 (95% CI, 0.748–0.872) and 0.831 (95% CI, 0.773–0.890); (C) the AUC of MRadiomics+MCNNs and MRadiomics+MCNNs+Clinical is 0.811 (95% CI, 0.749–0.873) and 0.834 (95% CI, 0.776–0.892); (D) the AUC of MClinical is 0.686 (95% CI, 0.617–0.756) and the lowest one. MRadiomics is less efficient than other models except MClinical. There is no significant different between MMCNNs, MRadiomics+MCNNs and MRadiomics+MCNNs+Clinical. MCCNs, multi-level residual CNNs; CNN, convolutional neural network; AUC, area under the receiver operating characteristic curve; EGFR, epithelial growth factor receptor.