Model "P" in gender prediction based on the mastoid process.

Aim To determine the degree of accuracy in determining the sex of the skull based on classical morphometric analysis of the mastoid process. Morphometric analysis excluded a subjective approach expressed in osteoscopic analysis. Methods The study was conducted on a sample of 100 macerated skulls of known gender and age from the second half of the 20th century, including the Bosnian population. Of the 100 skulls, 50 (50%) were male and 50 (50%) were female. Male skulls were on average 60 (47-71) years old and female 57 (43-76) years old. At each mastoid process, 3 measurements were made: mastoid length, mastoid width (medio-lateral diameter) ML) and anteroposterior diameter (AP) of the mastoid process. Results Using the univariate method, we found a significant difference between length, width, AP diameter and size of mastoid processes as well as between gender (p <0.05). Multivariate binary logistic regression showed statistically significant differences in AP diameter of the mastoid process (p<0.05). Conclusion The created model "P" ("P"=exp [X]/1+exp [X]) for sex determination based on mastoid process showed sensitivity of 82% correct prediction for female skulls and 65% accurate prediction for male skulls. This discourse with respect of population standards grants most effective anthropological proof and as such may be suggested for forensic expertise based on human skull.