Facial Landmarks Determination with Different Digital Scanners: An In Vivo Study

Background: The development of facial scanning technology has introduced new methods for facial morphology analysis, progressively shifting from conventional methodology such as direct anthropometry and two-dimensional photography toward three-dimensional digital acquisition. These technologies aim to reduce operator subjectivity, enhance measurement reproducibility, and enable comprehensive facial analysis within digital workflows. Methods: Thirty adult volunteers were recruited and provided informed consent. In each participant, twenty-five predefined facial landmarks were identified and nineteen linear interlandmark distances were recorded using three methods: direct anthropometric measurement with a digital caliper (Mitutoyo®, USA), a low-cost portable facial scanner (Revopoint®), and a professional static facial scanner (RAYFace®). Manual anthropometry was used as a clinical refence standard. All measurements were performed by a single trained operator. Trueness was defined as the absolute difference between the reference measurements and the mean of digital measurements, while precision was defined as the standard deviation of repeated digital measurements. Results were expressed as mean values and 95% confidence intervals. Results: Overall precision was 0.58 (0.53; 0.62) mm for Revopoint® and 0.43 (0.39; 0.47) mm for RAYFace®, corresponding to precision percentages of 1.19 (1.06; 1.33) % and 0.88 (0.78; 0.97) %, respectively. Mean trueness values were 2.16 (2.01; 2.31) mm and 1.92 (1.80; 2.05) mm for conventional-Revopoint® and conventional-RAYFace®, corresponding to a trueness impact value of 4.30 (3.87; 4.74) % and 4.08 (3.61; 4.55) %, respectively. Statistically significant differences between scanners were observed for specific landmark locations. Conclusions: Within the methodological limitations of this in vivo study—including the use of manual anthropometry as a reference standard, a single-operator design and a predominantly female sample—both facial scanners demonstrated reproducible linear measurements within clinically acceptable thresholds for prosthodontic and esthetic planning applications. The professional static scanner showed superior accuracy, suggesting greater suitability for complex cases requiring higher precision.