Radiomics Models Based on Magnetic Resonance Imaging for Prediction of the Response to Bortezomib-Based Therapy in Patients with Multiple Myeloma

Table 3

The comparison of different models.

Model	Training set				Validation set
	AUC	ACC	SEN	SPE	AUC	ACC	SEN	SPE
RF	0.896	0.868	0.867	0.870	0.885	0.833	0.923	0.727
KNN	0.840	0.755	0.800	0.696	0.832	0.792	0.923	0.636
SVM	0.856	0.792	0.800	0.783	0.818	0.708	0.769	0.636
LR	0.855	0.755	0.700	0.826	0.811	0.708	0.692	0.727
DT	0.853	0.830	0.733	0.957	0.808	0.792	0.769	0.818
Bayes	0.879	0.792	0.800	0.783	0.801	0.833	0.923	0.727

AUC: area under the curve; ACC: accuracy; SEN: sensitivity; SPE: specificity; RF: random forest; KNN: -nearest neighbor; SVM: support vector machine; LR: logistic regression; DT: decision tree.