Sex estimation by volumetric evaluation of the maxillary canine using cone-beam computed tomographic images

Abstract

Abstract Background Teeth can be a reasonable part of sex estimation in case of being the single evidence available or in doubt although they are not as accurate as bones in sex estimation. The size of teeth is greater in males than in females. In this study, it was aimed to determine cut-off values of the tooth volume (TV) and root volume (RV) of the maxillary canines and to evaluate the validity of the method for discriminating between males and females. Methods Cone beam computed tomography (CBCT) images of 250 individuals aged 18 to 63 years with equal age and sex distribution were assessed retrospectively. The sample divided into reference set including 168 images of 84 females and 84 males and validation set including 82 images of 48 females and 34 males. Receiver operating characteristic (ROC) curve analysis, and Youden’s index were used to determine cut-off values of the volumetric measurements and to test the discriminative performance of the method. Results The optimal cut-off values of TV ≥ 581 mm3 and RV ≥ 334 mm3 for estimating sex were established from the reference set. The sensitivity (Se), specificity (Sp), and accuracy (Ac) were 0.77 for TV ≥ 581 mm3 while they were 0.82, 0.77, and 0.79 for RV ≥ 334 mm3. When we tested the previously derived cut-off values of TV of the maxillary canine, the Se, Sp, and Ac were found to be respectively 0.71, 0.90, and 0.82 for TV ≥ 619 mm3 and 0.97, 0.27, and 0.56 for TV ≥ 510 / 460 mm3 (right/left). Conclusions The identified cut-off values of TV and RV of the maxillary canines can be used to estimate sex with moderate accuracy when the only evidence available is teeth or in case of doubt. The method’s applicability should be assessed for different populations.