Sex Prediction Using Geometric Morphometry with Parameters Obtained From Computed Tomography Images of the Seventh Cervical Vertebra

Authors : Yusuf Seçgin, Halil Şaban Erkartal, Zaid Ali Naji Zanqah, Nesibe Yılmaz, Nevin Köremezli Keskin, Şeyma Toy

Pages : 756-764

Doi:10.35440/hutfd.1779800

View : 55 | Download : 110

Publication Date : 2025-12-25

Article Type : Research Paper

Abstract :Background: The durability of vertebrae increases their likelihood of being discovered in forensic contexts or natural disasters. Identification procedures can also be facilitated through these structures. The aim of this study, based on this hypothesis, is to determine whether it is possible to identify gender with high accuracy and reliability using geometric morphometry on specific points identified on the 7th cervical vertebra. Materials and Methods: The study was conducted using computed tomography images from 300 individuals aged 18-65. The images were converted into three dimensions and superimposed. The images at actual magnification were then converted to TPS format and 30 landmarks were added. Generalized Procrustes Analysis was applied to these raw coordinates, and the coordinates were repositioned around the center of gravity. Principal Component Analysis was then applied to these repositioned data to reduce dimensionality. Linear discriminant analysis (LDA) was applied to the reduced-dimensionality images to obtain the gender prediction accuracy rate. Results: The study found that 63.674% of the coordinates clustered around the center of gravity could be explained by the first 6 principal components. LDA analysis applied to the new coordinates yielded an 83.33% gender prediction rate. Conclusions: The study achieved a high accuracy rate for gender prediction by applying geometric morphometry to points identified on the 7th cervical vertebra.
Keywords : Vertebra prominens, Geometrik morfometri, Bilgisayarlı tomografi, Cinsiyet tahmini

ORIGINAL ARTICLE URL