Comprehensive Maxillary Anterior Teeth Aesthetics Analysis With Deep Learning in Dental Anthropometry

The height and width of maxillary front teeth, their width-to-height ratio—W/H% (gold standard), their width ratio (gold proportion), smile width, buccal corridor, and marginal gingiva define smile form. The study examined the characteristics of dental students with complete sets of teeth and used deep learning in dental anthropometry. The study involved 88 subjects. The central incisor was measured in the subjects’ mouths, while the gold proportion was photometrically established. Clinical examinations assessed smile breadth, buccal corridor visibility, and gingival contour class. The Adaptive Neuro-Fuzzy Inference System (ANFIS) was employed for deep learning analysis because of its ability to model complex, non-linear relationships in high-dimensional data. The average length of the central incisor in men amounted to 10.5 mm, whereas the value in women was 10.2 mm. The central incisor length and W/H% did not differ significantly between men and women. In the majority of subjects, the gold proportion in the width of the front teeth was established and was uniform between men and women. On average, 8 to 10 teeth were visible during smiling in the subjects. There was no significant difference in smile width between men and women. A pronounced buccal corridor was present in half of the subjects, without a statistically significant difference between the sexes. The results showed a much higher prevalence of class 1 gingiva than class 2 gingiva, with no significant gender-related difference. The results of ANFIS analyses, represented by Root Mean Square Error (RMSE) values, showed non-linear relations among anthropometric parameters. The study found similarities between anthropometric measurements obtained from the Serbian subjects and findings previously reported in the literature. Modelling complex dental anthropometric relationships using ANFIS helps improve understanding of interactions among dental parameters.