Machine Learning Based on a Multiparametric and Multiregional Radiomics Signature Predicts Radiotherapeutic Response in Patients with Glioblastoma

Table 4

A summary of the AUC with different machine learning methods on the independent test dataset.

Algorithm	AUC	95% CI	SENS	SPEC	ACC	value
RF	0.937	0.813-0.989	0.829	1	0.9	<2.2-16
SVM	0.874	0.732-0.957	0.771	0.800	0.775	1.375-08
KNN	0.874	0.731-0.958	0.714	0.800	0.725	7.63-05
Logistic	0.931	0.805-0.987	0.886	1	0.900	<2.2-16
Xgboost	0.880	0.738-0.961	0.886	0.800	0.875	6.851-09

SENS: sensitivity; SPEC: specificity; ACC: accuracy.