Sex estimation from morphometric measurements of os coxae in computed tomography images and validity and accuracy of “diagnose sexuelle probabiliste V2” software for the Turkish population

This study aimed to analyze the reliability, applicability, and accuracy of the “Diagnose Sexuelle Probabiliste V2” software used for sex estimation on pelvic CT images of the Turkish population. Pelvic CT images of 200 individuals (100 male, 100 female) were used. Slicer 5.8.0 software was used to create a three-dimensional representation of the bone and perform the measurements. 10 morphometric measurements were made on the os coxae to estimate the sex of the individuals on the images. Of the 10 variables used in DSP2, the difference between the sexes was significant for all variables except SA (p < 0.001). When the posterior probability threshold was set at 0.95 or higher, the sex of only 3 individuals could not be estimated using the 10 variables in DSP2. The sexing success rate was 98.5%. The accuracy rate was 100%. DSP2 is a highly reliable method for sex estimation from pelvic CT images for the Turkish population.