Abstract

Background: Artificial intelligence (AI) has been a contribution in recent years to the development of new tools for dental, surgical, and esthetic treatment. In the case of image diagnosis, AI allows automated analysis of some facial parameters. The aim of this study was to evaluate the precision and reproducibility of these IA analyses compared with a human operator. Methods: Thirty subjects were selected, and frontal and profile photographs were taken of each. The photographs were previously calibrated and then analyzed by a human operator and an AI for vertical, horizontal, symmetry parameters, and facial profile angles. Results: Statistically significant differences were found in at least 1 vertical, horizontal, symmetry, and facial profile measurement. Conclusions: The 2D facial analysis performed on photographs using AI is an interesting and constantly evolving tool. However, there are still many differences with the analysis performed by the human operator, so it should be used carefully. Further studies using metric measurements should be performed to contrast the values obtained by human operators and AI and to analyze the accuracy of point selection in facial soft tissues. © © 2024 Mutaz B. Habal, MD. All rights reserved.