Fig. 6: An integrated diagnosis model for PTC.

a The heatmap of 20 DEGs identified by recursive feature elimination and logistic regression analysis for discriminating PTC and BTN. b Receiver operating characteristic (ROC) analysis for discriminating PTC from BTN based on Logistic Regression (LR), Support Vector Machine (SVM), Naïve Bayes (NB), Random Forest (RF) and K-Nearest Neighbor (KNN) algorithm in training cohort (left) and validation cohort (right).