Fig. 5From: CT features and quantitative analysis of subsolid nodule lung adenocarcinoma for pathological classification predictionGraph shows ROC curves predicted AAH/AIS and MIA value, Predicted probability, mean size, maximum diameter, mean CT value and maximum CT value, and the areas under the curve for the identification of MIA were 0.815, 0.729, 0.728, 0.733, and 0.761, respectively The area under the curve, sensitivity and specificity were higher than for the single factor analysis indicators of various quantitative parametersBack to article page