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

Table 5

AKI classification performance using clinical-based random forest (RF) model.

Signature	AUC	Precision	Recall	-score	Accuracy	Delong test for AUC values
Training set
Fold-1	0.957	0.714	0.909	0.800	0.889	,
Fold-2	0.934	1.000	0.308	0.471	0.800
Fold-3	0.927	1.000	0.636	0.778	0.909
Fold-4	0.881	1.000	0.462	0.632	0.848
Fold-5	1.000	1.000	1.000	1.000	1.000
Average	0.940	0.943	0.663	0.736	0.889
Testing set
Fold-1	0.686	0.471	0.296	0.364	0.691	,
Fold-2	0.665	0.522	0.231	0.320	0.718
Fold-3	0.583	0.267	0.074	0.116	0.665
Fold-4	0.659	0.485	0.308	0.376	0.706
Fold-5	0.714	0.571	0.500	0.533	0.767
Average	0.661	0.463	0.282	0.342	0.709
Best model in validation set	AUC	Precision	Recall	-score	Accuracy	Delong test for AUC values
	0.817	0.620	0.677	0.647	0.788