Clinical decisions in Orthodontics using x-ray-based images and artificial intelligence approaches: a scoping review.

INTRODUCTION

Artificial intelligence (AI) in health has increased its applications over the last years. The large amount of data available due to the improvement of data storage and digital exams provided better knowledge in treatment planning and opened new possibilities of applications of AI in Orthodontics' diagnosis and treatment planning.

OBJECTIVE

The aim of this scoping review was to examine when AI models enhance the clinical decision-making process in orthodontic diagnosis and treatment planning, with a focus on the utilization of X-ray-based imaging.

METHODS

Individual eletronic search strategies were developed and conducted on PubMed/Medline, Scopus, Web fo Science, Embase, Lilacs, and Cochrane Library (only English language articles published from January 2000 to October 20, 2021), aiming for relevant studies that met the eligibility criteria.

RESULTS

12 studies were included, categorized in 5 different groups: Orthognathic surgery, Temporomandibular Joint (TMJ) osteoarthritis, skeletal pattern, obstructive sleep apnea (OSA), and skeletal maturation/development. Most of the AI models used were Deep Learning (DL) based and the X-ray image that was most used was lateral radiographs.

CONCLUSION

Integrating AI into clinical practice will likely continue to evolve, enhancing treatment planning in selective cases. The best applications were over TMJ osteoarthritis, skeletal maturation, and classification, OSA, and the need for orthognathic surgery.