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

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Orthodontics

Background:

• Three-dimensional facial anthropometry is crucial in orthodontics and orthognathic surgery.
• Conventional systems (e.g., Vectra®, 3dMD®) are accurate but expensive and complex.
• The MetiSmile® scanner presents a cost-effective, portable alternative requiring validation.

Purpose of the Study:

• To rigorously evaluate the accuracy and reproducibility of the MetiSmile® 3D face scanner.
• To compare digital measurements with direct anthropometry.
• To assess intra- and inter-observer reliability.

Main Methods:

• Two-phase validation on a mannequin head.
• Phase 1: Mesh reproducibility under varying lighting and refinement (AL, n-AL, AL-F) using Iterative Closest Point (ICP).
• Phase 2: Direct Anthropometry (DA) vs. Digital Anthropometry (DiA) measurements by three observers, assessing reliability with intraclass correlation coefficients (ICCs).

Main Results:

• Phase 1 showed high mesh reproducibility with mean root-mean-square error (RMS) as low as 0.030 mm (AL-F).
• Largest deviations observed near facial features like eyes and lips.
• Phase 2 demonstrated excellent ICCs (≥0.997) and minimal mean absolute DA-DiA differences (0.25-0.33 mm).

Conclusions:

• The MetiSmile® scanner produces highly reproducible 3D facial meshes.
• Linear measurements obtained are clinically acceptable for mannequin models.
• Further validation on live subjects is necessary for routine clinical adoption.