A Novel Radiomics-Based Machine Learning Framework for Prediction of Acute Kidney Injury-Related Delirium in Patients Who Underwent Cardiovascular Surgery

Table 6

The AKI classification performance using clinical-based support vector machine (SVM) model.

Signature	AUC	Precision	Recall	-score	Accuracy	Delong test for AUC values
Testing set
Fold-1	0.535	0.973	0.777	0.864	0.729	,
Fold-2	0.500	1.000	0.767	0.868	0.696
Fold-3	0.558	0.861	0.786	0.822	0.700
Fold-4	0.548	0.938	0.777	0.850	0.724
Fold-5	0.500	1.000	0.777	0.874	0.713
Average	0.528	0.954	0.777	0.856	0.712
Training set
Fold-1	0.612	0.247	1.000	0.396	0.124	,
Fold-2	0.769	0.865	0.776	0.818	0.867
Fold-3	0.650	0.938	0.818	0.874	0.844
Fold-4	0.808	0.849	0.776	0.811	0.889
Fold-5	0.615	0.938	0.776	0.849	0.778
Average	0.731	0.882	0.778	0.826	0.849
Best model in validation set	AUC	Precision	Recall	-score	Accuracy	Delong test for AUC values
	0.812	0.821	0.742	0.780	0.870