Principal component analysis of 3-dimensional facial soft-tissue morphology in three adult populations

This preliminary study aims to identify soft tissue facial morphological features that vary among ethnically diverse, facially balanced adults using Principal Component Analysis (PCA) of 3D facial scans. A total of 210 3D facial scans (35 per subgroup: Chinese, Hungarian, and Hispanic; stratified by sex) were selected from a database at the University of Alabama. Facial scans were captured using 3D-laser scanning and stereophotogrammetry (3dMD). A total of 57 landmarks were manually placed on each scan. The landmark coordinates were analyzed using Generalized Procrustes Analysis and PCA (conducted in R) to identify principal components contributing most significantly to shape variation. PCA identified four principal components (PCs) accounting for 77.72% of the total variance in soft tissue morphology. PC1 (49.13%) was associated with upper facial height. PC2 (17.70%) reflected the spatial relationship between nose protrusion and eye position. PC3 (6.31%) corresponded to interocular distance and vertical eye placement. PC4 (4.14%) represented upper lip protrusion. These PCs showed that the greatest morphological variability was in the upper facial region. Observed differences were interpreted within clinical and aesthetic contexts and were consistent with findings from prior literature on facial development and attractiveness. Significant variations in upper facial soft tissue morphology exist among normal adult subjects from different ethnic backgrounds. This study demonstrates the utility of PCA in revealing clinically relevant patterns in facial morphology and highlights the importance of individualized, ethnically sensitive treatment planning in orthodontics and orthognathic surgery. Future longitudinal and AI-driven studies are recommended to refine personalized diagnostics and develop inclusive aesthetic standards.