Evaluation of Facial Anthropometry Using Three-Dimensional Photogrammetry and Direct Measuring Techniques

In this study, 14 standard facial distances on 10 adult volunteers were measured directly with a caliper and indirectly on two-dimensional images using the 3dMDtrio system. Two raters performed the measurements with at least 1 week between rating sessions. The intra- and inter-rater reliabilities and agreement of the measurements were calculated using intra-class correlation coefficient (ICC), mean absolute difference (MAD), and Bland-Altman plots with limits of agreement (LOA). The 2 raters had an average discrepancy (MAD) of 1.6 mm when their digital measurements were compared to their direct measurements. The reliability of the digital and direct methods varied greatly depending on which of the 14 anthropometric distances that was being assessed. Only 6 digitally and 5 directly measured anthropometric distances showed both an ICC >0.75 and a MAD <1 mm, in the intra-rater as well as the inter-rater measurements. The Bland-Altman plots and LOA displayed the same pattern. In summary, the digital and direct methods were generally compatible in terms of reliability and agreement. However, the reliability and agreement between the 14 anthropometric measurements varied considerably, indicating that poor landmark identification is the main limitation to both modern and traditional measuring techniques in the face. Consequently, some anthropometric landmarks warrant further definition or prior anthropometric training by the evaluators. The authors also recommend that the MADs and LOAs provided in this report are put into relation to the facial distance that is being evaluated and its clinical context.