Sex prediction from the femur and hip bone using a sample of CT images from a Spanish population.

Sex estimation and the analysis of sexual dimorphism is an essential part of forensic and archaeological studies of skeletons. However, osteologists often have to rely on single measurements, such as femoral head diameters, to estimate sex, especially when skeletons are incomplete. We have obtained a sex-prediction model based on CT images by applying the logistic regression technique to the measurements obtained for the proximal femoral epiphyses and coxal. Nine variables for 114 Spaniards (58 females and 56 males) of known age and sex from a region close to Madrid have been studied. The prediction equation obtained using these nine variables correctly classifies 99.1 % of these individuals. Reducing the equation to the three most explanatory variables (VDH, HDH and MIB) resulted in the correct classification of 98.3 %. These findings suggest that this procedure is highly effective for sex prediction. However, a lack of expertise may produce biases in the measurements obtained from CT images. Moreover, these equations are only most effective for the population for which they were calculated as human growth and body size are sensitive to nutritional variations, environmental stress and the so-called secular trend.