A clinical–radiomics model based on noncontrast computed tomography to predict hemorrhagic transformation after stroke by machine learning: a multicenter study

Insights into Imaging

Table 2 Performance of five methods to select clinical features in the training cohort

Selecting method	Dim	AUC	ACC	SEN	SPE	PPV	NPV
LASSO	15	0.900	0.822	0.859	0.792	0.776	0.870
SelectFromModel	10	0.892	0.829	0.834	0.825	0.800	0.856
RFECV	15	0.915	0.826	0.856	0.798	0.781	0.871
RFE	5	0.917	0.832	0.815	0.846	0.818	0.844
LR	13	0.903	0.827	0.842	0.814	0.793	0.860

LASSO, Least Absolute Shrinkage and Selection Operator; RFECV, Recursive Feature Elimination Cross Validation; RFE, Recursive Feature Elimination; LR, Logistic Regression; Dim, dimension; AUC, area under the receiver operator characteristic curve; ACC, accuracy; SEN, sensitivity; SPE, specificity; PPV, positive predictive value; NPV, negative predictive value