Table 6.

Comparison of the accuracy of age estimation between different regression models for males and females in the test dataset 2 (120 OPGs).

Gender	Number	Formula	Difference in ME (SD)	P-value	Difference in MAE (SD)	P-value	Difference in RMSE
Males	60	PT + TM vs. PT	0.50 (1.49)	0.01*	−0.80 (1.20)	0.00*	−0.75
		PT + TM vs. TM	0.03 (0.59)	0.71	−0.07 (0.44)	0.22	−0.02
		PT vs. TM	−0.47 (1.64)	0.03*	0.73 (1.25)	0.00*	0.73
Females	60	PT + TM vs. PT	0.15 (1.69)	0.48	−0.87 (1.30)	0.00*	−1.05
		PT + TM vs. TM	−0.19 (0.69)	0.03*	−0.23 (0.63)	0.01*	−0.24
		PT vs. TM	−0.35 (2.00)	0.18	0.65 (1.54)	0.00*	0.81
Total	120	PT + TM vs. PT	0.32 (1.59)	0.03*	−0.84 (1.25)	0.00*	−0.89
		PT + TM vs. TM	−0.08 (0.65)	0.17	−0.15 (0.54)	0.00*	−0.12
		PT vs. TM	−0.41 (1.82)	0.02*	0.69 (1.40)	0.00*	0.77

PT: permanent teeth 31–37; TM: third molars; ME: mean error; MAE: mean absolute error; SD: standard deviation; RMSE: root mean square error.

P-value: obtained using paired samples t test or Wilcoxon's signed rank test.

^*Values that showed significant difference.