Table 5.

Results (in F₁ score) of teeth recognition after applying candidate optimization algorithm for tenfold cross validation.

	Number of teeth in test dataset	ResNet-50			ResNet-101
		Original	ωc = 0.8 ωp = 0.2	ωc = 0.5 ωp = 0.5	Original	ωc = 0.8 ωp = 0.2	ωc = 0.5 ωp = 0.5
K1	2947	0.957	0.969	0.971	0.975	0.976	0.977
K2	2890	0.962	0.967	0.967	0.967	0.969	0.970
K3	2904	0.961	0.969	0.969	0.970	0.973	0.974
K4	2909	0.962	0.969	0.969	0.967	0.972	0.972
K5	2958	0.974	0.980	0.979	0.982	0.984	0.984
K6	2847	0.958	0.969	0.969	0.971	0.975	0.974
K7	2862	0.974	0.983	0.984	0.984	0.987	0.987
K8	2796	0.956	0.961	0.963	0.964	0.967	0.967
K9	2846	0.951	0.962	0.964	0.962	0.968	0.970
K10	2969	0.965	0.976	0.975	0.978	0.983	0.983
Average	2893	0.962	0.971	0.971	0.972	0.975	0.976

Bold values indicate the best results in that particular row (particular section).