Table 2.

Validation effect of different metrics on three validation datasets

	Class	Sensitivity	Specificity	Pos Pred value	Neg Pred value	F1	Balanced accuracy
I	H	0.95	0.9	0.89	0.95	0.92	0.92
	L	0.75	0.9	0.68	0.93	0.71	0.83
II	H	0.91	0.8	0.91	0.8	0.91	0.86
	L	0.77	0.99	0.83	0.98	0.8	0.88
III	H	0.77	0.82	0.74	0.84	0.76	0.8
	L	0.8	1	1	0.98	0.89	0.9

H predicted as high blood Phe concentration, L predicted as low blood Phe concentration