Fig. 6From: Virtual biopsy using CT radiomics for evaluation of disagreement in pathology between endoscopic biopsy and postoperative specimens in patients with gastric cancer: a dual-energy CT generalizability studyCalibration curves and decision curve analysis of the nomogram. Calibration curves of the nomogram showed good agreement between the predicted and observed pathologic disagreement probability in both the training (A) and test set (B). The Hosmer–Lemeshow test yielded a nonsignificant statistic (p-value = 0.115, and 0.226), suggesting there is no significant departure. Decision curves analysis (C) for training set and validation set indicated that when the nomogram is used to predict the risk (probability) of disagreement pathological status, patients could obtain better clinical benefits within wide range of risk (probability) threshold in both sets. The gray curve represents the hypothesis that all pathological status were discordant. The black line represents the hypothesis that no disagreement pathological statusBack to article page