Fig. 3

HCC-risk nomogram construction and validation. (a) Risk nomogram including GPC3 and GLS1 for predicting HCC. (b) Calibration curves depict the calibration of the nomogram in terms of the agreement between the predicted risk of HCC and actual HCC. The 45˚ blue line represents a perfect prediction, and the red dashed line represents the predictive performance of the nomogram, together with a bias-corrected black solid line. The closer the dashed line to the ideal line, the better the predictive accuracy of the nomogram. (c) Decision curve analysis for the HCC-risk nomogram. The red line represents the nomogram, the gray line represents the hypothesis that all patients have HCC, and the black line represents the hypothesis that no patients have HCC. The decision curve shows that if the threshold probability is between 0.54 and 1, then using the HCC-risk nomogram to predict HCC status adds more benefit than taking intervention to either all or no patients