The 82nd Annual Meeting of the American Association of Physical Anthropologists (2013)


The reality of virtual anthropology: testing the utility of computer generated models for the quantitative assessment of the cranium

AMBER D. WHEAT and BRIDGET FB. ALGEE-HEWITT.

Anthropology, University of Tennessee

Saturday All day, Clinch Concourse Add to calendar

Recent advances in data acquisition methods for geometric morphometry offers opportunities for improving upon traditional procedures for craniofacial analysis by changing the way in which we access skeletal materials and evaluate population-informative traits. We argue that a virtual skeletal collection containing three dimensional models of the cranium would (1) improve ease of access to reference materials, (2) encourage experiential equality among osteologists in terms of the quantity and variation of materials analyzed, and (3) provide a wealth of study materials in a “virtual” bone lab. Multiple laser scans of individual cranium from six Danish cemetery populations in the Anthropological Data Base Odense University (ADBOU) skeletal collection were taken using the Next Engine Desktop Scanner. This high-definition color copy of each cranium allows for the study of the skeletal features in a fully-resolved three-dimensional space. Prior research shows promise for replacing the actual skeletal materials with virtual models. We confirm the quality of these models by comparing craniometric data obtained from the laser scans against coordinate data collected in the laboratory setting using a Microscribe digitizer and a subset of the actual crania. We demonstrate the unique advantages of virtual models in quantifying non-metric cranial traits by means of sliding semi-landmarks and coordinate meshes, using the open-access software, Meshlab and tpsDig/tpsUtil. We offer recommendations for three-dimensional model building using laser scans, coordinate extraction from these virtual crania, and the effective use of this high dimensional data for studies of human variation and for individual identifications in forensic contexts.

This study was funded by a National Science Foundation Dissertation Improvement Grant awarded to Bridget Algee-Hewitt (BCS-676917).

Tweet
comments powered by Disqus