Accuracy of MRI skeletal age estimation for subjects 12–19. Potential use for subjects of unknown age

• Published in: International journal of legal medicine Vol. 129; no. 3; pp. 609 - 617
• Main Authors: Serinelli, Serenella, Panebianco, Valeria, Martino, Milvia, Battisti, Sofia, Rodacki, Karina, Marinelli, Enrico, Zaccagna, Fulvio, Semelka, Richard C., Tomei, Ernesto
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.05.2015; Springer Nature B.V
• Subjects: Accuracy; Adolescent; Age; Age Determination by Skeleton - methods; Carpal Bones - anatomy & histology; Child; Criminal law; Female; Forensic Medicine; Growth Plate - anatomy & histology; Hand Bones - anatomy & histology; Humans; Magnetic Resonance Imaging - methods; Male; Males; Medical Law; Medicine; Medicine & Public Health; Observer Variation; Original Article; Osteogenesis - physiology; Reference Values; Reproducibility of Results; Young Adult; Rome Italy; Italy; Forensic; Age estimation; Hand–wrist; Children; MRI
• ISSN: 0937-9827; 1437-1596; 1437-1596
• DOI: 10.1007/s00414-015-1161-y

Introduction In forensic practice, there is a growing need for accurate methods of age estimation, especially in the cases of young individuals of unknown age. Age can be estimated through somatic features that are universally considered associated with chronological age. Unfortunately, these features do not always coincide with the real chronological age: for these reasons that age determination is often very difficult. Our aim is to evaluate accuracy of skeletal age estimation using Tomei’s MRI method in subjects between 12 and 19 years old for forensic purposes. Material and methods Two investigators analyzed MRI images of the left hand and wrist of 77 male and 74 female caucasian subjects, without chronic diseases or developmental disorders, whose age ranged from 12 to 19 years. Skeletal maturation was determined by two operators, who analyzed all MRI images separately, in blinded fashion to the chronological age. Inter-rater agreement was measured with Pearson ( R 2 ) coefficient. One of the examiners repeated the evaluation after 6 months, and intraobserver variation was analyzed. Bland–Altman plots were used to determine mean differences between skeletal and chronological age. Results Inter-rater agreement Pearson coefficient showed a good linear correlation, respectively, 0.98 and 0.97 in males and females. Bland–Altman analysis demonstrated that the differences between chronological and skeletal age are not significant. Spearman’s correlation coefficient showed good correlation between skeletal and chronological age both in females ( R 2  = 0.96) and in males ( R 2  = 0.94). Conclusions Our results show that MRI skeletal age is a reproducible method and has good correlation with chronological age.