Skip to main content
Fig. 4 | BMC Cancer

Fig. 4

From: A combined nomogram based on radiomics and hematology to predict the pathological complete response of neoadjuvant immunochemotherapy in esophageal squamous cell carcinoma

Fig. 4

Comparison of predictive performance of the combined nomogram, radiomics signature, and hematology model. A: Receiver operating characteristic curves of the combined nomogram, radiomics signature, and hematology model. B: Nomogram based on independent predictors (radScore, hemScore, cN stage, and tumor length). C: Calibration curves of the combined nomogram to estimate the consistency among the estimated pCR probability by the combined nomogram and the authentic pCR. The ideal situation is shown by the black dashed line, which acts as the reference line and shows when the predicted and actual values coincide; the actual situation of the nomogram is shown by the solid blue line, referred to as the apparent line; the bias-corrected line is shown by the solid green line, which shows the corrected nomogram's actual situation. D: The models' clinical benefits were assessed using decision curve analysis. The black horizontal line indicates that when all patients do not receive treatment, regardless of the probability threshold, there is no clinical net benefit. The gray diagonal represents the change in clinical net benefit as the probability threshold changes when all patients receive treatment

Back to article page