Skip to main content
. 2022 Jul 3;14(13):3264. doi: 10.3390/cancers14133264

Figure 2.

Figure 2

Workflow. The workflow is composed of 3 main phases, “Feature selection Phase”, “Model selection Phase”, and “Test Phase”. ” Feature selection Phase” includes determination of appropriate feature number and feature selection. “Model selection Phase” includes parameter determination with nested validation and evaluation of the learning model. “Test Phase” is a final test of models with 10 independent cases for final validation. * Test cases. X: 7 for GdT1, 2 for T2. Abbreviations: recursive feature elimination—RFE; area under the curve—AUC; logistic regression—LR; support vector machine—SVM.