Virtual anthropology? Reliability of three-dimensional photogrammetry as a forensic anthropology measurement and documentation technique

• Published in: International journal of legal medicine 2021-05, Vol.135 (3), p.939-950
• Main Authors: Omari, Rita, Hunt, Cahill, Coumbaros, John, Chapman, Brendan
• Format: Article
• Language: English
• Subjects: Accident reconstruction; Anthropology; Biomedical materials; Body measurements; Computed tomography; Correlation coefficients; Forensic anthropology; Forensic engineering; Forensic Medicine; Image reconstruction; Medical imaging; Medical Law; Medicine; Medicine & Public Health; Model accuracy; Morphology; Original Article; Photogrammetry; Reliability; Scanning; Skull; Statistical tests; Three dimensional models; Variance analysis
• ISSN: 0937-9827; 1437-1596
• DOI: 10.1007/s00414-020-02473-z

Osseous remains provide forensic anthropologists with morphological and osteometric information that can be used in building a biological profile. By conducting a visual and physical examination, an anthropologist can infer information such as the sex and age of the deceased. Traditionally, morphological and osteometric information is gathered by physically handling remains for analysis. With the advancement of digital technology, there has been a shift from direct to indirect methods of analysis by utilizing models generated from three-dimensional (3D) imaging, which includes computed tomography (CT) scanning and 3D photogrammetry. Although CT scanning is more common, photogrammetry has found application in a range of fields such as architecture, geography and road accident reconstruction. The application of modern-day photogrammetry for forensic anthropology purposes, however, has not been discussed extensively. The aim of this research was to validate the accuracy of 3D models generated by photogrammetry by comparing them to both 3D models generated by CT scanning and the actual physical models. In this study, six 3D models were created using photogrammetry ( n  = 3) and CT scanning ( n  = 3). The 3D models were generated from three different Bone Clone® human skulls. A mobile phone camera was used to capture images, which were then processed in Agisoft Metashape®. Intrarater, interrater, and intermethod reliability tests gave correlation coefficients of at least 0.9980, 0.9871, and 0.9862, respectively; rTEM results ranged from 0.250 to 6.55%; and an analysis of variance (ANOVA) yielded P values under 0.05 for all measurements except one. Statistical tests therefore showed photogrammetry to be a reliable and accurate alternative to more expensive CT scanning approaches.