Fig. 6: Construction and validation of a nomogram integrating MPMRecNet with clinicopathological variables. | npj Digital Medicine

Fig. 6: Construction and validation of a nomogram integrating MPMRecNet with clinicopathological variables.

From: Deep learning-enabled multiphoton microscopy predicts colorectal cancer recurrence from routine FFPE specimens

Fig. 6

a Univariable logistic regression analysis of clinical features and the MPMRecNet prediction score. b Multivariable logistic regression identifying independent predictors of recurrence. c Nomogram model constructed using independent predictors to estimate individualized recurrence risk. d Calibration curve of the nomogram model, showing agreement between predicted and observed recurrence rates. e ROC curve comparison of the nomogram, MPMRecNet, and individual clinical variables in the external validation cohort. f Decision curve analysis comparing the net clinical benefit of the nomogram, MPMRecNet, and individual predictors across varying threshold probabilities.

Back to article page