Mapping The Future: Algorithm Predicts Female Adolescent CVM Stages

The present study was designed to define a novel algorithm capable of predicting female adolescents’ cervical vertebrae maturation stage with high recall and accuracy. A total of 560 female cephalograms were collected, excluding those with unclear vertebral shapes or deformed scales. 480 films from female adolescents (mean age: 11.5 years; range: 6–19 years) were used for model development, while 80 subjects were randomly and stratified into a validation cohort. Predictive parameters from 15 anatomic points and 25 quantitative parameters of the second to fourth cervical vertebrae (C2-C4) were used to establish the ordinary logistic regression model. Evaluation metrics including precision, recall, and F1 score were employed to assess the efficacy of the models in each identified cervical vertebrae maturation stage (iCS). Four key parameters (chronological age, D3:AH3, @4, and C3lp-C4up) were integrated into the ordinary regression model, achieving 94.00% accuracy, 93.98% precision, 93.98% recall, and 93.95% F1-score. Despite the hybrid logistic-based model achieving high accuracy, the unsatisfactory performance of stage estimation was noticed for the third stage (CS3) in the primary cohort (89.20%) and validation cohort (85.00%). Through bivariate logistic regression analysis, PH4 was further selected in CS3 to establish a corrected model, thus the evaluation metrics were upgraded to 95.83%, and 96.64%, respectively. Our novel logistic model generated stage-specific formulas and demonstrated exceptional performance, establishing its potential as a benchmark for maturity assessment in clinical craniofacial orthopedics for Chinese female adolescents.