1Department of Anthropology, University of Manitoba, 2Department of Forensic Medicine, University of Copenhagen
Thursday All day, Plaza Level
Increasingly, the use of 3D coordinate data is being used for a variety of osteological analyses, including sex determination. Most studies make use of a digitizer or laser scanner to capture such data (especially for cranial landmarks), but increasingly computed tomography (CT) is being utilized as the basis for such analyses. This paper presents on the results of 3D shape analysis of cranial morphology for determining sex from human skeletal remains. 3D landmark data were collected from the orbits, mandible and skull of approximately 100 individuals from a documented age and sex CT dataset derived from post-mortem scans to investigate determination of sex. Segmentation and rendering of the CT data was done using Materialise MIMICS medical imaging software to create 3D models of the skull for each individual.
To investigate shape independent of size for each area of the skull, the coordinate data were analysed using the morphometric statistical software program PAST (PAleontological STatistics). A procrustes fit was used to separate size and shape by scaling the raw coordinate data. Discriminant function analysis was undertaken to assess the relative accuracy of sex determination in this sample for each area of the skull from the 3D coordinate data. High levels of overall accuracy were observed (>90%) for discriminating sex from the shape data. The results are evaluated relative to other methods of sex determination in the skeleton, and the practical considerations of using a 3D coordinate approach from CT data are discussed.