Comparing a common clavicle maturation-based age estimation method to ordinary regression analyses with quadratic and sex-specific interaction terms in adolescents

Sebastian R. Reder,Isabel Fritzen,Marc A. Brockmann,Jochen Hardt,Katrin Elsner,Katja Petrowski,Monika Bjelopavlovic
DOI: https://doi.org/10.1038/s41598-024-52980-x
IF: 4.6
2024-02-03
Scientific Reports
Abstract:Established methods of age estimation are based on correlating defined maturation stages of bony structures with tables representing the observed range of biological ages in the majority of cases. In this retrospective monocentric study in southwestern Germany, common age estimation methodology was assessed in n = 198 subjects at the age of 25 or younger by analyzing the influence of age, quadratic age, biological sex and age-sex interaction on the ossification stages of the medial epiphysis fugue. Three readers (ICC ≥ 0.81 for left/right side) evaluated routine care computed tomography images of the clavicle with a slice thickness of 1 mm. By using least square regression analyses, to determine the real biological age a quadratic function was determined corrected for the age estimated by established methods and sex (R 2 = 0.6 each side), reducing the mean absolute error and root mean squared error in the age estimation of women (2.57 and 3.19) and men (2.57 and 3.47) to 1.54 and 1.82 for women, and 1.54 and 2.25 for men. In women, the medial clavicle epiphysis seem to fuse faster, which was particularly observable from approximately 18 years of age. Before that age, the estimation method was relatively close to the ideal correlation between assessed and real age. To conclude, the presented new method enables more precise age estimation in individuals and facilitates the determination and quantification of additional variables, quantifying their influence on the maturation of the medial clavicle epiphysis based on the established ossification stages.
multidisciplinary sciences
What problem does this paper attempt to address?