Use of Artificial Intelligence and Manual Method for Cephalometric Analysis in Lateral Skull Radiographs

Palomino K.P.P.; Condori L.A.; Velazco G.C.V. · Smart Innovation, Systems and Technologies (2026) · DOI: 10.1007/978-3-032-24262-4_22

Resumen

Cephalometry is a widely used tool in orthodontics, and the use of artificial intelligence tools could improve its efficiency and accuracy, making it crucial to compare it with the traditional method to determine its clinical applicability. The present study aims to evaluate similarities and differences in the use of artificial intelligence and the manual method when performing cephalometric analyses on lateral skull radiographs. The research is quantitative, with a non-experimental and cross-sectional design. The study considered a total of 200 lateral skull radiographs of patients of both sexes aged between 14 and 54 years. To evaluate the variables, Steiner's cephalometric analysis was performed using Webceph Version 2.0.0 AI software, while the manual method was performed conventionally. The data were processed with SPSS v.22 and Excel v.2016 and analyzed with the Wilcoxon statistic because they were related groups. According to the results obtained by comparing measurements made using artificial intelligence and the manual method, statistically significant differences were found in the measurements of SNA, SNB, ANB, OCCLUSAL PLANE/SN, U1-NA (º), L1-NB (mm), and L1-NB (º). On the other hand, no differences were observed in the measurements of Go-Gn.SN, U1-NA (mm), and interincisal. The study concludes by demonstrating a statistically significant difference when performing Steiner's analysis using artificial intelligence compared to the manual method. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.

Instituciones: Universidad Nacional del Altiplano