Table 2.

Mean error (ME), mean absolute error (MAE), mean square error (MSE), root mean square error (RMSE), and R² values assessing performance of machine learning regression methods and Cameriere European/Chinese formula for chronological age estimation

Method	ME ± SD	MAE ± SD	MSE ± SD	RMSE ± SD	R2 ± SD
Linear regression	0.008 ± 0.052 (− 0.095–0.094)	0.553 ± 0.026 (0.501–0.589)	0.488 ± 0.063 (0.396–0.588)	0.698 ± 0.045 (0.629–0.767)	0.909 ± 0.012 (0.890–0.925)
Support vector machine	0.004 ± 0.063 (− 0.142–0.104)	0.489 ± 0.030 (0.422–0.552)	0.392 ± 0.049 (0.286–0.480)	0.625 ± 0.039 (0.535–0.693)	0.925 ± 0.011 (0.900–0.949)
Random Forest	− 0.004 ± 0.046 (− 0.090–0.088)	0.495 ± 0.024 (0.446–0.533)	0.389 ± 0.039 (0.309–0.461)	0.623 ± 0.032 (0.556–0.679)	0.928 ± 0.009 (0.914–0.945)
European formula	0.592 ± 0.032 (0.532–0.654)	0.846 ± 0.228 (0.801–0.891)	0.755 ± 0.038 (0.684–0.829)	0.869 ± 0.022 (0.827–0.911)	–
Chinese formula	0.386 ± 0.035 (0.322–0.450)	0.812 ± 0.022 (0.530–0.655)	0.890 ± 0.049 (0.796–0.997)	0.943 ± 0.026 (0.892–0.999)	–