A fundamental understanding of the correlation between cephalometric measurements is essential for advancing our comprehension of the anatomy of the oral and maxillofacial regions. Given the potential for discrepancies in input data due to the analyst's choices, it is crucial to minimize bias and establish a reliable standard for malocclusion research. To this end, a comprehensive range of variables was employed to investigate the correlation structure of the cephalometric measurement variables.

Furthermore, in order to conduct data-driven analyses of various malocclusions, it is essential to establish a reference point based on studies of normal occlusion. The study was performed using data from 735 adults aged 18-25 years with normal occlusion.

Network analysis can be utilized to examine the complex correlation structure between a multitude of variables. This structure can be elucidated through the application of weighted network analysis and minimum spanning trees. This analytical approach allows for the identification of the structure of clusters and the determination of the core structure of the correlation between cephalometry variables.

It is proposed that an investigation of the correlation between cephalometric variables through network analysis may significantly enhance our understanding of the anatomical characteristics of the oral and maxillofacial regions. Furthermore, this will provide an important foundation for studying malocclusion using artificial intelligence based on data.