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

Table 3

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

Algorithm	AUC	95% CI	SENS	SPEC	ACC
RF	0.980	0.942-0.996	0.946	0.932	0.939	<2.2-16
SVM	0.965	0.921-0.988	0.905	0.972	0.939	<2.2-16
KNN	0.969	0.926-0.990	0.757	0.932	0.844	<2.2-16
Logistic	0.974	0.933-0.993	0.932	1	0.966	<2.2-16
Xgboost	0.962	0.917-0.987	0.865	0.905	0.885	<2.2-16

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