Analysis of carpal bones on MR images for age estimation: First results of a new forensic approach

Current multifactorial age estimation methods are based on radiography, however, in the forensic field there is growing interest in using magnetic resonance imaging (MRI). With regard to the carpal region, MRI provides more information for defining the individual ossification nuclei and the cartilage surrounding single bones. During the phase of bone growth, the progressive reduction of the cartilage layer is accompanied by the development of a cartilage-bone interface. The aim of our study was to create a new model for age estimation, based on the ratio between the area occupied by the nucleus of ossification (NO) and the surface of growth (SG) of each carpal bone, the latter derived by adding NO to the area of cartilage-bone interface. We analyzed 57 MRI scans of Italian subjects aged between 12 and 20 years, without growth diseases, endocrine disorders or osteodystrophy. Measurements of NO and SG areas were extracted using ImageJ software, and the ratio between the NO and SG of each bone (NOSG) was calculated. A multiple linear regression model was used to estimate the individual's age as a function of the variables: gender and wrist bone measurements. The results showed that the best model was obtained with 6 predictors (nvmax=6): Gender, and the NOSG of the Trapezoid, Trapezium, Scaphoid, Pisiform, and Capitate. The median of the residuals (observed age minus predicted age) was -0.025 years, with an IQR of 0.19 years. Thus a new forensic approach to age assessment using MRI is introduced in this paper, which gives the preliminary results.