Predicting micropapillary or solid pattern of lung adenocarcinoma with CT-based radiomics, conventional radiographic and clinical features

Table 4 model evaluation measurements

Name	AUC	AUC_upper95%CI	AUC_lower95%CI	Accuracy	Kappa	Sensitivity	Specificity	Pos_Pred_Value	Neg_Pred_Value	F1
rad_train	0.863	0.817	0.909	0.785	0.569	0.782	0.787	0.789	0.780	0.785
rad_test	0.849	0.774	0.924	0.762	0.521	0.720	0.800	0.766	0.759	0.742
cli_train	0.771	0.713	0.713	0.695	0.391	0.597	0.795	0.747	0.660	0.664
cli_test	0.778	0.687	0.868	0.705	0.404	0.620	0.782	0.721	0.694	0.667
comb_train	0.872	0.829	0.916	0.785	0.569	0.758	0.811	0.803	0.767	0.780
comb_test	0.853	0.782	0.925	0.790	0.577	0.700	0.873	0.833	0.762	0.761

ISSN: 1465-993X