Table 2 The fitting effect of different models.

From: Boosting algorithm improves the accuracy of juvenile forensic dental age estimation in southern China population

Method

Gender

CAa

DAb

ADc

P valued

MAE (< 18)

MAE (< 16)

Demirjian

Female

11.18(3.83)

11.49(3.72)

0.31(1.18)

0.00***

0.95

0.94

Demirjian

Male

10.68(3.76)

11.10(3.74)

0.42(1.15)

0.00***

0.96

0.96

Willems

Female

11.18(3.83)

11.02(3.78)

 − 0.16(0.98)

0.00***

0.78

0.73

Willems

Male

10.68(3.76)

10.77(3.70)

0.09(0.98)

0.02***

0.77

0.76

Modified

Female

11.18(3.83)

11.16(3.73)

 − 0.02(0.97)

0.61

0.77

0.71

Modified

Male

10.68(3.76)

10.68(3.55)

0.01(0.96)

0.83

0.76

0.72

  1. MAE Mean absolute error.
  2. aChronological age.
  3. bDental age.
  4. cAge deviation, AD = dental age–chronological age. Chronological age, dental age and the age deviation are given as the mean (standard deviation).
  5. dPaired T test significance of chronological age and dental age.