Digital vs. Direct Anthropometry with MetiSmile® 3D Face Scanner: A Validation and Reliability Study on a Mannequin Model.

BACKGROUND: Three-dimensional facial anthropometry is increasingly used in orthodontics and orthognathic surgery. Conventional face scanning systems such as Vectra® and 3dMD® are well validated but remain costly and technically demanding. The MetiSmile® 3D face scanner provides a more affordable and portable alternative, yet its accuracy and reproducibility have not been rigorously evaluated. METHODS: Validation was performed on a standardized mannequin head in two phases. Phase 1 assessed mesh reproducibility under artificial lighting (AL), natural lighting (n-AL), and after mesh-refinement (AL-F). Landmark-based pre-registration with Iterative Closest Point refinement was applied; root-mean-square error (RMS) and distance maps were calculated. In phase 2, three observers (student, resident, consultant) performed 14 predefined linear measurements by direct anthropometry (DA) and digital anthropometry (DiA). Intra- and inter-observer reliability was evaluated using intraclass correlation coefficients. RESULTS: Phase 1 yielded mean RMS values of 0.041 mm (AL), 0.043 mm (n-AL), and 0.030 mm (AL-F), with largest deviations near eyes, alar regions, and lip commissures. Phase 2 showed excellent ICCs (≥0.997) and mean absolute DA-DiA differences of 0.25-0.33 mm, with only few differences > 2 mm. CONCLUSION: The MetiSmile® scanner generates highly reproducible meshes and clinically acceptable linear measurements on mannequin models. Further validation on live subjects is warranted before routine clinical application.