Abstract: Objective: To establish and explore a predictive model for hand wrist bone age calculation based on two-dimensional (2D) and three-dimensional (3D) morphological measurements of cervical vertebra in cone beam computerized tomography (CBCT) images. Methods: Hand-wrist X-ray images and craniofacial CBCT images of 150 Eastern Chinese female adolescents (aged from 8 to 16 years) were collected, and 2D and 3D morphological variables from cervical vertebras were measured. According to Fishman's Skeletal Maturity Index, the stages of bone age were determined. Multiple stepwise regression analysis was used to develop two different bone age predictive formulas, excluding and including 3D measurements. The paired t-test was applied to detect the prediction reliability of the models. The adjusted R-squared and accuracy of the models excluding and including 3D measurements were analyzed and compared. Results: Age and quantitative variables of the second, third and fourth cervical vertebrae were positively correlated with bone age. The paired t-test showed that there was no significant difference between the predicted results of the two bone age formulas and the gold standard (P >0.05). The adjusted R-squared and accuracy of the formula including 3D measurements were improved compared to the formula which 3D measurements were excluded. Conclusions: The adjusted R-squared and accuracy of the hand wrist bone age predictive model formulated were improved by conjunctive using of 2D and 3D morphological measurements, which can provide a reliable index for quantitative craniomaxillofacial growth studies in eastern Chinese female adolescents, and also a reference for individual bone maturity.

Key words: [Key words]: hand wrist, cervical vertebra, bone age, 3D measurements, eastern China