Comparison of Different 3D Surface Registration-Based Methods to Assess Facial Asymmetry

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Craniofacial Analysis

Background:

• Facial asymmetry assessment is increasingly important in clinical diagnostics.
• Existing 3D surface-based methods for facial asymmetry may yield non-equivalent data.
• A comparative analysis of these methods is lacking.

Purpose of the Study:

• To compare three distinct methods for assessing facial asymmetry.
• To evaluate asymmetry of the entire face and its thirds (upper, middle, lower).
• To determine if observed differences between methods are clinically acceptable.

Main Methods:

• Utilized two maxillofacial (trigeminal nerve-based) and one orthodontic (reference plane-based) method on 3D facial models of 80 healthy adults.
• Calculated facial asymmetry using Root Mean Square values via surface-based registration.
• Assessed method reliability (intra- and inter-operator), differences (ANOVA), interchangeability (Bland-Altman, Similarity Percentage), and time efficiency.

Main Results:

• All methods exhibited excellent intra- and inter-operator reliability.
• Significant differences in asymmetry measurements were found between methods (p < 0.001).
• Differences were clinically acceptable (<0.50 mm) between trigeminal methods for all regions, and between orthodontic and trigeminal methods for the whole face and upper third. The orthodontic method was faster.

Conclusions:

• No single method is universally superior for facial asymmetry assessment; interchangeability is not guaranteed across all methods.
• Trigeminal subdivision-based methods are interchangeable for morpho-functional analysis of maxillofacial regions with diverse embryological origins.
• Clinical context dictates method selection, emphasizing the importance of comparing data with studies using demonstrated interchangeable methods.