A Genomic-Clinicopathologic Nomogram for the Prediction of Lymph Node Invasion in Prostate Cancer

Figure 5

Performance of SVM models in imbalanced and SMOTE-balanced datasets. (a) ROC curves of the SVM model in imbalanced training and validation sets. (b) ROC curves of the SVM model in SMOTE-balanced training and validation sets. (c) The predictive performance of SVM models in imbalanced and SMOTE-balanced training sets. (d) The predictive performance of SVM models in imbalanced and SMOTE-balanced validation sets. ROC: receiver operating curve; AUC: area under the ROC curve; NPV: negative predictive value; PPV: positive predictive value; SVM: support vector machine; SMOTE: synthetic minority oversampling technique.

(a)

(b)

(c)

(d)