Stima dell’età ossea sul vivente a scopo forense mediante analisi dei nuclei di ossificazione del carpo su Imaging a Risonanza Magnetica

In recent years, scientific surveys for age estimation have greatly expanded their horizons, from the cadaver to the living. Regarding the living, this kind of verification is required for different areas: in adoption procedures, in the context of sporting competitions, about immigration, in the protection of unaccompanied minors, for forensic purposes such as for child offense or in the ascertainment of the crime of child pornography. In providing the scientific data, the forensic physician must consider variability factors related to ethnicity or race, environmental conditions, diversity of skeletal maturation of each individual, pre-existing pathologies. Hence the difficult task of providing the result in the most easily intelligible way for the judge, possibly in a numerical way and with statistical indications, highlighting the margin of error and cut-off. In the growing subjects the most used indicators for the forensic age determination are linked to the evaluation of sexual maturity, skeletal maturity and dental maturity. Although there are no univocally recognized procedures, currently the most used methods include a physical examination and anthropometric analysis, an orthopantomographical analysis and a left hand and wrist x-ray study. For minors, age assessment evokes ethical issues, with regard to the use of radiation as well as recommended by many operational protocols. The international organizations for the protection of minors have underlined in more circumstances that it can not be considered ethically correct to expose a minor to x-ray analysis in the absence of clinical reasons. In this scenario, several recent studies have investigated the possible use of exams in the field of age determination that do not provide ionizing radiation, such as Nuclear Magnetic Resonance (NMR). The use of NMR on regions investigated through x-ray exams is linked to the need to compare the results with existing data for age estimation. This is the reason why the wrist and the left hand, as well as the sternal extremity of the clavicle, have been investigated using NMR techniques. Anatomical regions that are poorly visualized by x-ray exams, such as the distal tibial and calcaneal epiphysis, as well as the proximal tibial epiphysis and distal femoral, were also investigated. The analysis of the carpal region by NMR for age estimation has been presented in a few works, some of which have considered the development of carpal bones together with the maturation of ulna and radius. The eight carpal bones undergo an endochondral ossification process, each from a single center; capitate is the first whereas the pisiform is the last to appear and it begins to ossify at the age of 9-10 years in females and 12 years in males. The order of ossification of the other bones is variable and depending on sex, food, race; ossification is complete only between 20 and 24 years. The progressive growth of the carpal ossification nucleus in adolescents is accompanied, both by a progressive reduction of the external cartilage layer and a reduction of the interface area located between the bone nucleus and the external cartilage. The aim of our project was to create a model for the age estimation based on the ratio between the area occupied by the nucleus of ossification (NO) and the surface of growth deriving from the NO + the interface area (SG). An observational retrospective study was carried out in collaboration with the Pediatric and Specialistic Radiology Departement of AOU - Ospedali Riuniti di Ancona: 40 Italian subjects were selected (21 females and 19 males) in a range between 12 and 20 years, undergoing left-hand NMR in the period 2012-2018. The subjects were free from any bone growth disorders. The instrument used was the same for all subjects: a 1.5 T open NMR (Philips Achieva). The NMR image of each carpal bone, extrapolated with the help of a radiologist, it was analyzed using the ImageJ software, through which the NO and the SG were selected. This procedure was repeated for each carpal bone and the full data were saved in a file in excel format. Once the NO and the SG of each bone were obtained, we calculated the ratio between the sum of all the NO (NOtot) and the sum of all SG (SGtot) of the all 8 bones (NOSGtot). This procedure was repeated for each of the 40 subjects. The intra and inter-operator variability was estimated; the results attesting a good reproducibility of the method. The results obtained allowed to realize a linear regression model with the following equation: age= α+β∙〖NOSG〗_tot+γ∙sex+ε Where sex = 1 for male, 0 for female and  is the random error with normal distribution, mean 0 and variance 2. The parameters of the model, ,  and , were estimated using the least squares method. The estimated standard error was 0.28 years. The value of the regression determination coefficient was R2 = 0.988, indicating an optimal adaptation of the model to the data. The method will be tested on a wider number of cases and on different ethnic groups. In combination with other techniques, the model developed may be applied for forensic purposes in the future.