Related Experiment Video
Updated: Sep 14, 2026
Reliability of Artificial Intelligence-Based Cone Beam Computed Tomography Integration with Digital Dental Images
Published on: February 23, 2024
Augmented and mixed reality as auxiliary tools in dental procedures and their integration with artificial
Matheus Pinós1, Newton Sesma2, Giovani Flach1
1Department of Dentistry, São Leopoldo Mandic University, São Paulo, Brazil.
Objectives:
To map and synthesize the scientific evidence regarding the use of Augmented Reality (AR) and Mixed Reality (MR) as auxiliary tools in dental procedures, and to examine the extent to which Artificial Intelligence (AI) has been integrated into these workflows.
Data:
A scoping review was conducted in accordance with the PRISMA-ScR guidelines and the Joanna Briggs Institute (JBI) recommendations. The study selection and data extraction processes were carried out by two independent reviewers, with disagreements resolved by consensus.
Sources:
Searches were performed in PubMed, Scopus, and Web of Science for studies published up to June 2025.
Study Selection:
Fifteen studies were included from a total of 1,457 records identified. Most of the evidence was concentrated in implant dentistry, where MR-based dynamic navigation achieved sub-millimetric accuracy (e.g., three-dimensional entry deviation of 0.417 mm) and outperformed the freehand technique. Improvements in angular accuracy (up to 72% among inexperienced operators) were observed in orthodontic mini-implant placement, and more conservative tissue removal was reported in endodontic access and in tooth preparation. In zygomatic implant placement, however, the freehand technique was not surpassed in apical or angular accuracy, and accuracy improved only when robotic stabilization was added. Artificial Intelligence was identified in a minority of the included studies and, where present, was restricted to preparatory stages of the workflow, namely automated segmentation, image registration, and planning support, rather than to intraoperative execution.
Conclusions:
AR and MR were applied predominantly as intraoperative visualization and navigation aids and showed accuracy superior to freehand execution, whereas AI functioned as an enabling technology confined to the preparatory stages of the workflow. Given the predominance of in vitro and proof-of-concept designs, these findings indicate technical feasibility rather than established clinical effectiveness, and multicenter randomized trials with independent validation of the systems remain necessary before clinical adoption can be established.
