Table 1.

Performance comparison of single-parametric internal and external classifiers.

ROI	Parameter used	AUC	Sensitivity	Specificity	Accuracy	Statistical significance
Internal	NAK	0.75	0.72	0.81	0.77	***
	ENT	0.76	0.82	0.69	0.76	***
	CONV	0.75	0.70	0.81	0.76	***
	CONH	0.74	0.70	0.80	0.75	***
	CORV	0.64	0.67	0.75	0.71	**
	CORH	0.64	0.72	0.68	0.70	**
	ENEV	0.65	0.67	0.69	0.68	**
	ENEH	0.62	0.58	0.76	0.67	*
	HOMV	0.64	0.61	0.73	0.67	**
	HOMH	0.65	0.82	0.58	0.70	**
	VARV	0.70	0.47	0.93	0.71	***
	VARH	0.59	0.53	0.81	0.67	non-significant
External	NAK	0.81	0.98	0.61	0.79	***
	ENT	0.83	0.72	0.92	0.82	***
	CONV	0.82	0.81	0.81	0.81	***
	CONH	0.68	0.95	0.46	0.70	***
	CORV	0.75	0.79	0.73	0.76	***
	CORH	0.55	0.28	0.93	0.61	non-significant
	ENEV	0.83	0.84	0.76	0.80	***
	ENEH	0.78	0.82	0.73	0.78	***
	HOMV	0.86	0.72	0.93	0.83	***
	HOMH	0.69	0.70	0.73	0.72	***
	VARV	0.81	0.82	0.81	0.82	***
	VARH	0.74	0.82	0.68	0.75	***

Statistical significance is divided into classes based on p-values and marked as follows: non-significant ($p\ge 0.05$), *($p<0.05$), **($p<0.01$) and ***($p<0.001$).