Age determination in forensic medicine by examination of pelvis and proximal femoral epiphyses using computed tomography images

Age determination is a current research topic in clinical forensic medicine. Internationally standardized approaches are being established through scientific investigations, and national population-specific age determination studies are being done. In national studies, it is known that the atlases have a high standard error in certain age groups, necessitating the investigation of complementary methods in addition to traditional age determination techniques. In the postmortem period, since age determination is carried out by analyzing bone remnants, it is considered beneficial to present age determination methods specific to anatomical regions. In our study, we aimed to contribute to the literature and develop an age determination method specific to the Turkish population by examining ossification centers and the degree of fusion between sacral vertebrae in pelvis and proximal femur computed tomography (CT) images in the 5-30 age range. Abdominal-pelvic CT scans of cases aged 5-30 years, conducted by the Department of Radiology at Hacettepe University Faculty of Medicine between June 1, 2020, and August 16, 2023, were reviewed. After applying exclusion criteria, a total of 260 cases were randomly selected for the study. In the abdominal-pelvic CT scans, three ossification centers in the proximal femur, five in the coxal bone, and six regions in the sacrum were examined in all three planes and staged according to their closure/fusion status. The correlation between ossification centers, sacral fusion stages, and age was analyzed using the Spearman Correlation Test. Polynomial regression and step-wise methods were applied to evaluate the usability of independent variables in age determination, and models were created. The suitability of polynomial regression models for age estimation was evaluated using the coefficient of determination (R2), p-value, and Standard Error of Estimation. The Cohen-kappa (κ) non-parametric test was used to assess inter-observer and intra observer consistency. The correlation analysis between variables and age revealed a very strong correlation between the iliac crest and age, a moderate correlation between the left ischiopubic synchondrosis and S1-1 intervertebral fusion and age, and a strong correlation between other ossification centers, costal process fusion level, and age. It was determined that all cases with iliac crest stage 3 were 17 years of age or older, all male cases with left triradiate ossification center stage 3 were 12 years of age or older, and all male cases with left tuber ischiadicum ossification center stage 2 were 15 years of age or older. In the ossification centers in the proximal femur, cases with left femoral head stage 3 were found to be 12 years of age or older in males and 13 years of age or older in females; cases with left greater trochanter stage 3 were 14 years of age or older in males and 13 years of age or older in females; cases with left lesser trochanter stage 3 were 14 years of age or older in males and 13 years of age or older in females. As a result, it was understood that centers other than ischiopubic synchondrosis and S1-2 intervertebral could be used as additional methods for forensic purposes. It is determined that that previously obtained abdominal-pelvic CT images taken for medical indications in living individuals can be used for age determination, and in postmortem cases, abdominal-pelvic CT scans performed during postmortem analysis can be used for determining the biological profile.