Fusion of FDG-PET Image and Clinical Features for Prediction of Lung Metastasis in Soft Tissue Sarcomas

Figure 4

Performance comparison of feature selection. (a) represents the overall performance from the perspective of all data sets. (b–d) denote the performance from the three types of data sets including training, validation, and test set, respectively. Feature_selection represents 24 features selected by random forest and test methods. Original denotes 67 features including 48 image features and 16 clinical features. Only Radiomics represents 48 image features, and Only Clinical denotes 16 clinical features.

(a)

(b)

(c)

(d)