Table 3.

Performance of assigning patients in each verification cohort to the four clusters generated by k-means on the main cohort, Cohort-1.

	Cohort-1A			Cohort-1B			Cohort-2			Cohort-3			Cohort-4			Mean
	ACC	SEN	SPE	ACC	SEN	SPE	ACC	SEN	SPE	ACC	SEN	SPE	ACC	SEN	SPE	ACC	SEN	SPE
MHO	0.905	0.992	0.850	0.948	0.892	0.981	0.854	0.692	0.962	0.942	0.875	1.000	0.924	0.894	0.953	0.914	0.869	0.949
HMO-U	0.875	0.780	0.927	0.908	0.908	0.908	0.831	0.956	0.765	0.934	1.000	0.904	0.941	0.941	0.942	0.898	0.917	0.889
HMO-I	0.954	0.677	0.986	0.972	0.731	1.000	0.962	0.667	1.000	1.000	1.000	1.000	0.983	0.615	1.000	0.974	0.738	0.997
LMO	0.962	0.755	1.000	0.980	1.000	0.974	0.985	0.944	0.991	0.993	0.950	1.000	0.965	1.000	0.963	0.977	0.930	0.986
Mean	0.924	0.801	0.941	0.952	0.883	0.966	0.908	0.815	0.929	0.967	0.956	0.976	0.953	0.862	0.964	0.941	0.863	0.955

ACC, Accuracy; HMO-I, hypermetabolic obesity hyperinsulinemia subtype; HMO-U, hypermetabolic obesity hyperuricemia subtype; LMO, hypometabolic obesity; MHO, metabolic healthy obesity; SEN, Sensitivity; SPE, Specificity.