Resumen

D’ANGELO DEL CAMPO, Manuel Domingo et al. Intermethod error in geometric morphometric and the relevance of texturization and landmark marking. Rev Arg Antrop Biol [online]. 2023, vol.25, n.1, e057.  Epub 01-Ene-2023. ISSN 1514-7991.  http://dx.doi.org/10.24215/18536387e057.

Morphometric analyses lead to biases in the level of precision, and therefore, error. Geometric morphometrics development has made it possible to create digital collections composed by records from diverse sources. The combined use of data obtained through multiple methods introduces a new type of error, the inter-method error. The joint use of distinct digitization sources will result in low error, as long as there are no significant differences in the results obtained among techniques. An analysis of the Procustes variance, a principal component analysis on the Procrustes coordinates and a hierarchical cluster analysis were carried out to analyse the inter-method and intra-observer error in eight human skulls from Southern Patagonia, digitized by computerized tomography, surface scanner and photogrammetry, using 35 landmarks type I, II and III. The results show that there are no significant differences between the digitization sources, so 3D files from different sources could be used together. No significant intra-observer error was observed for any of the sources, also presenting a lower magnitude than the inter-method error. In the present study, photogrammetry, the only method that permits recovering the texture and in which the landmarks were previously indicated, is the source that presents the lowest error. Based on the results obtained, it is suggested that it is possible to perform satisfactory geometric morphometric analyses regardless of the source used for its registration, considering those analysed here, highlighting the influence of the texture and the registration of landmarks on the degree of error.

Palabras clave : bioanthropology; virtual anthropology; Southern Patagonia.

        · resumen en Español     · texto en Español     · Español ( pdf )