A network approach to orthodontic diagnosis

BACKGROUND: Network analysis, a recent advancement in complexity science, enables understanding of the properties of complex biological processes characterized by the interaction, adaptive regulation, and coordination of a large number of participating components. OBJECTIVE: We applied network analysis to orthodontics to detect and visualize the most interconnected clinical, radiographic, and functional data pertaining to the orofacial system. MATERIALS AND METHODS: The sample consisted of 104 individuals from 7 to 13 years of age in the mixed dentition phase without previous orthodontic intervention. The subjects were divided according to skeletal class; their clinical, radiographic, and functional features were represented as vertices (nodes) and links (edges) connecting them. RESULTS: Class II subjects exhibited few highly connected orthodontic features (hubs), while Class III patients showed a more compact network structure characterized by strong co-occurrence of normal and abnormal clinical, functional, and radiological features. Restricting our analysis to the highest correlations, we identified critical peculiarities of Class II and Class III malocclusions. CONCLUSIONS: The topology of the dentofacial system obtained by network analysis could allow orthodontists to visually evaluate and anticipate the co-occurrence of auxological anomalies during individual craniofacial growth and possibly localize reactive sites for a therapeutic approach to malocclusion.