Fig. 5From: Predicting micropapillary or solid pattern of lung adenocarcinoma with CT-based radiomics, conventional radiographic and clinical featuresRadscores and cliscores in training (a,b) and test (c,d) set. ROC curves (e,f) and calibration curves (g,h) of three models. In ROC curves, higher AUC indicated better capability of discrimination. In calibration curves, being closer to the 45-degree black dotted line indicated better model fittingBack to article page