Sex determination by pelvic osteometry using discriminant analysis

BACKGROUND: One of the current challenges of forensic medicine is the sex identification of skeletal remains. Sex determination from isolated bones is considered a particularly difficult task. Consequently, increased attention has been paid to investigating the possibility of sex identification from an isolated pelvis.

AIM: To develop a method for morphological sex determination by pelvis osteometry using multiple discriminant analysis and evaluating its accuracy.

MATERIALS AND METHODS: The objects of the study were the pelvic skeletons (n=84) of individuals of both sexes aged 18 to 86 years from the collection of the Department of Human Anatomy of the V.I. Razumovsky Saratov State Medical University, as well as 11 male (19–76 years) and 14 female (18–80 years) pelvises from the collection of the Department of Human Anatomy, Operative Surgery and Topographic Anatomy of the St. Luke Lugansk State Medical University. Measurements were made by V.P. Alekseev’s method using a large spreading caliper, a sliding caliper, and a measuring tape. Classification models were built by stepwise discriminant analysis.

RESULTS: A model with a set of variables is obtained, which allows determining the sex of skeletal remains with consideration of osteometric data. The identification accuracy of the model was 90.4%. When tested on the verification group, an accuracy of 84% was obtained. The following data used in the mathematical model showed statistically significant differences: pelvis height, ischium height, and the largest diameter of the acetabulum (p <0.01).

CONCLUSION: The study allowed creating a mathematical model that can be used for sex determination of skeletal remains taking into account seven osteometric parameters of the pelvis; the theoretical and practical accuracy of the model was 90.4% and 84%, respectively.