Skip to main content
. 2020 Jul 8;20:45. doi: 10.1186/s40644-020-00320-3

Fig. 5.

Fig. 5

a) A radiomics nomogram incorporating clinical features and a radiomics signature was developed in the training dataset. Calibration curves of the radiomics nomogram being used in the training dataset (b) and external validation dataset (c). The y-axis represented the actual lung adenocarcinoma rate, and the x-axis represented the predicted lung adenocarcinoma possibility. The diagonal dashed line indicated the ideal prediction by a perfect model