Table 4.

Performance of the Five Models in the Training and Test Sets

	Model	AUC (95% CI)	Specificity	Accuracy	Sensitivity
	Combined model	0.97 (0.936–1)	0.972	0.954	0.864
	Radiomics model	0.903 (0.845–0.961)	0.872	0.847	0.727
Training set	Clinical model	0.66 (0.55–0.77)	0.651	0.634	0.545
	BISAP model	0.62 (0.504–0.736)	0.578	0.588	0.636
	MRSI model	0.804 (0.714–0.894)	0.725	0.748	0.864
	Combined model	0.979 (0.947–1)	0.854	0.877	0.857
	Radiomics model	0.831 (0.723–0.939)	0.75	0.772	0.889
Test set	Clinical model	0.743 (0.59–0.896)	0.75	0.702	0.444
	BISAP model	0.668 (0.51–0.826)	0.667	0.649	0.556
	MRSI model	0.731 (0.537–0.926)	0.75	0.754	0.778

Abbreviations: AUC, area under the curve; BISAP, bedside index for severity in acute pancreatitis; MRSI, magnetic resonance severity index.