3D analysis of computed tomography (CT)-derived lumbar spine models for the estimation of sex.

Abstract

When skeletal remains are found scattered or in fragmentary conditions, the establishment of a biological profile of unknown individuals can be proven difficult. Consequently, multiple methods to ascertain the sex of the individual must be developed. The purpose of this study was to demonstrate that computed tomographic (CT)-derived 3D models of lumbar vertebrae could capture the unique morphologies of all five lumbar vertebrae to create equations for sex identification. The models were selected from a modern population consisting of 154 males and females that measured 30 standard linear measurements, the vertebral body wedging angle, and five aspect ratios. These measurements were then used to develop discriminant function equations for sex identification. Each lumbar level was analyzed individually as well as part of the entire lumbar spinal column. The results of this study showed that L1-L5 vertebrae can be used in sex determination with an 81.2-85.1% accuracy. When all five vertebrae are used in conjunction, the accuracy is 92.2%. The accuracy of the sex estimation found in this study for all lumbar vertebrae reinforces the distinct dimorphism between sexes while also providing forensic practitioners with more options or tools for their analyses.