Development of a personalized diagnostic model for kidney stone disease tailored to acute care by integrating large clinical, demographics and laboratory data: the diagnostic acute care algorithm - kidney stones (DACA-KS)

Table 4 Comparison of prediction performance of different models, using 10-fold cross validation

Model/Outcome		AUC	Sensitivity	Specificity	J
kidney stones (KS) vs. other genitourinary diseases (GUD)
(a)	Demographic	0.63 (0.02)	0.69	0.51	0.20
(b)	Charlson’s comorbidity index	0.69 (0.02)	0.69	0.62	0.31
(c)	eGFR	0.62 (0.02)	0.65	0.56	0.21
(d)	ICD	0.74 (0.02)	0.75	0.63	0.38
(e)	Labs	0.76 (0.02)	0.67	0.74	0.41
(f)	All	0.81 (0.02)	0.75	0.76	0.51
(g)	All (Stepwise)	0.80 (0.02)	0.76	0.71	0.47
(h)	STONE	0.62 (0.02)	0.55	0.64	0.19
KS vs. other conditions (OTH)
(a)	Demographic	0.65 (0.02)	0.64	0.62	0.27
(b)	CCI	0.65 (0.02)	0.68	0.57	0.25
(c)	eGFR	0.71 (0.02)	0.59	0.75	0.35
(d)	ICD	0.82 (0.02)	0.68	0.87	0.55
(e)	Labs	0.90 (0.01)	0.81	0.87	0.68
(f)	All	0.92 (0.01)	0.90	0.80	0.70
(g)	All (Stepwise)	0.92 (0.01)	0.90	0.81	0.71
(h)	STONE	0.64 (0.02)	0.62	0.65	0.27
KS vs. acute localized pain (ALP)
(a)	Demographic	0.60 (0.02)	0.71	0.47	0.18
(b)	Charlson’s comorbidity index	0.62 (0.02)	0.60	0.61	0.21
(c)	eGFR	0.57 (0.02)	0.59	0.44	0.15
(d)	ICD	0.77 (0.02)	0.74	0.69	0.43
(e)	Labs	0.85 (0.02)	0.66	0.90	0.56
(f)	All	0.88 (0.01)	0.78	0.85	0.63
(g)	All (Stepwise)	0.86 (0.01)	0.81	0.82	0.63
(h)	STONE	0.61 (0.02)	0.58	0.60	0.18

ISSN: 1472-6947