Skip to main content
. 2022 Aug 19;12:14132. doi: 10.1038/s41598-022-18085-z

Figure 6.

Figure 6

Results for the radiomic and deep feature-based models without including clinical data. (Left panel) Average AUCs on the five training splits. (Right panel) Average AUCs on the test splits. SFM SelectFromModel, PCA principal component analysis, CLUSTER feature agglomeration through clustering, SVM support vector machines, BAG bagging, RF random forest, XGB extreme gradient boosting, NNET neural network, NN k-nearest neighbours.