Development of Artificial Intelligence Models for Tooth Numbering and Detection: A Systematic Review

Area of Science:

• Dentistry
• Medical Imaging
• Artificial Intelligence

Background:

• Dental radiography is a crucial tool in dental practices and a valuable resource for advancing artificial intelligence (AI) technology.
• AI is increasingly explored for applications within dental radiology, particularly for image analysis and interpretation.

Purpose of the Study:

• To systematically review and critically appraise the development and performance of AI models for tooth detection and numbering using dento-maxillofacial radiographic images.
• To assess the diagnostic accuracy and evidence certainty of AI applications in dental radiology.

Main Methods:

• A systematic review adhering to PRISMA-DTA diagnostic test accuracy criteria.
• Comprehensive electronic searches across multiple databases (PubMed, Scopus, Embase, Cochrane, Web of Science, Google Scholar, Saudi Digital Library) for articles published between 2018 and 2023.
• Risk of bias assessment using QUADAS-2 and certainty of evidence evaluation using the GRADE approach for 16 selected articles.

Main Results:

• AI has been primarily applied to automated tooth detection and numbering, including identification in Cone-Beam Computed Tomography (CBCT) images and dental treatment pattern recognition.
• The reviewed AI models achieved a highest precision of 99.4% for tooth detection and 98% for tooth numbering.
• The certainty of evidence for AI applications in this domain was assessed using the GRADE approach.

Conclusions:

• AI technology demonstrates high accuracy and potential for automated tooth detection and numbering in dental radiology.
• AI can serve as a valuable supplementary diagnostic tool, enhancing the capabilities of dental radiology.
• Further research and validation are supported by the promising performance metrics of current AI models.