Related Experiment Video
Updated: Jan 15, 2026
Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
Published on: February 23, 2024
Comparison Between Direct, Virtual Aided by Clinician and Artificial Intelligence Bonding Techniques in Orthodontics
Tarek ElShebiny1, Johana Cortés-Mercado1, Neda Stefanovic2
1Department of Orthodontics, Case Western Reserve University, Cleveland, Ohio, USA.
Objectives:
The purpose of this study was to determine if there are any clinically significant differences between direct, virtual indirect, and artificial intelligence (AI) bonding techniques.
Materials And Methods:
This in vivo study analysed 840 teeth selected from 14 patients undergoing orthodontic treatment with full fixed appliances. Anatomical superimpositions were performed, and data were collected as both numerical values and colour-coded deviation maps to assess the differences between direct, AI, and virtual indirect bonding techniques.
Results:
The intraclass correlation coefficient test showed good correlation (0.894). The Kruskal-Wallis comparison showed a statistically significant difference when comparing direct to virtual indirect and direct to AI. Descriptive statistics showed 4 values with clinically significant differences when comparing direct to virtual indirect. Descriptive statistics showed 3 values with clinically significant differences when comparing direct to AI. Root mean square (RMS) discrepancies exceeding 0.50 mm were found in four tooth types (AI vs. Direct) and three (Clinician vs. Direct).
Conclusions:
We found statistically and clinically significant differences between AI and virtual indirect when compared to direct bonding. With our data, we could infer that if we compare AI versus virtual indirect, there might not be any clinically significant differences since the differences between them fall below 0.25 mm.
