Applying an automated method of classifying lip morphological traits

Abstract

Objective: To apply an automated computerised method to categorise and determine the prevalence of different types of lip traits, and to explore associations between lip traits and sex differences. Design: Observational descriptive study utilising an automated method of facial assessment. Setting and participants: A total of 4747 children from the Avon Longitudinal Study of Parents and Children (ALSPAC) who each had 3D facial scans carried out at 15 years of age. Methods: Each of the participants was automatically categorised regarding predetermined lip morphological traits. Descriptive statistics were applied to report the prevalence of the different types of each trait, and chi-square tests were used to investigate sex differences and associations between traits. Results: A total of 4730 individuals were assessed (47% male, 53% female). Eight predetermined lip traits have been reported previously. There were differences in prevalence for all lip traits in male and female patients (all P ⩽ 0.0002), with differences between the sexes described for each trait. For example, a deeply grooved philtrum of average width was more prevalent in boys, and an indentation near the upper vermilion border was more prevalent in girls. Each of the traits was significantly associated with the other traits (all P < 0.0001), with particularly strong associations seen between traits in the same region (e.g. upper lip). Individual associations between traits are reported; for example, a straight lip contour was found to be associated with no true vermilion border in both the upper and lower lip regions. Conclusion: The automated computerised method described is an invaluable tool for the categorisation of lip morphological traits. The prevalence of various types of traits has been described. Sexual dimorphism exists for all the lip traits assessed. Generally, each of the traits are associated with all other traits, with individual associations reported.