Skip to main content
. 2021 May 26;11:659969. doi: 10.3389/fonc.2021.659969

Figure 4.

Figure 4

Decision curve analysis (DCA) was conducted to evaluate the clinical usefulness of the models in prediction by quantifying the net benefits at different threshold probabilities in the training and testing set. The SVM model had the best performance in prediction of low- and high-grade renal cell carcinoma. Logistic, logistic regression model; SVM, support vector machine model; Decision tree, decision tree model.