Usefulness of an artificial intelligence–assisted indirect bonding method for optimizing orthodontic bracket positioning

Abstract

Objectives: To compare the bracket positioning accuracy of a traditional and an artificial intelligence (AI)-assisted digital indirect bonding (IDB) method to explore the current usefulness of AI for optimizing orthodontic bracket positioning. Materials and Methods: Twenty-five clinicians positioned brackets using traditional and AI-assisted digital IDB methods. Bracket positioning differences were quantified using digital superimposition of bracket setups and compared with an optimal setup. A total of 1800 bracket positioning differences were evaluated. One-tailed t-tests were used to determine whether these differences were within limits of 0.5 mm in mesial-distal and occlusal-gingival dimensions and within 2° for tip. Results: Overall mean bracket position differences between the traditional and digital setups were 0.28 mm for mesial-distal placement and 0.32 mm for occlusal-gingival placement; both were significantly below the 0.5-mm limit. In contrast, differences in tip were 3.4°, which was significantly greater than the 2° limit. Comparisons with an optimal setup showed overall statistically significant differences in mean bracket positioning for tip but not for the mesial-distal or occlusal-gingival mea-surements for both the traditional and AI-assisted digital IDB methods. However, the digital method was more accurate for bracket tip. Conclusions: Bracket positioning is consistent and highly accurate in linear dimensions with both traditional and digital IDB methods; however, AI may be useful for improving accuracy of bracket angulation. Clinicians who currently use traditional IDB methods may adopt AI-assisted digital IDB without compromising bracket positioning accuracy.