Dental implant planning using artificial intelligence: A systematic review and meta-analysis

Abstract

Statement of problem: Data on the role of artificial intelligence (AI) in dental implant planning is insufficient. Purpose: The purpose of this systematic review with meta-analysis was to analyze and evaluate articles that assess the effectiveness of AI algorithms in dental implant planning, specifically in detecting edentulous areas and evaluating bone dimensions. Material and methods: A systematic review was conducted across the MEDLINE/PubMed, Web of Science, Cochrane, and Scopus databases. In addition, a manual search was performed. The inclusion criteria consisted of peer-reviewed studies that examined the accuracy of AI-based diagnostic tools on dental radiographs for dental implant planning. The most recent search was conducted in January 2024. The Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool was used to assess the quality of the included articles. Results: Twelve articles met the inclusion criteria for this review and focused on the application of AI in dental implant planning using cone beam computed tomography (CBCT) images. The pooled data indicated an overall accuracy of 96% (95% CI=94% to 98%) for the mandible and 83% (95% CI=82% to 84%) for the maxilla in identifying edentulous areas for implant planning. Eight studies had a low risk of bias, 2 studies had some concern of bias, and 2 studies had a high risk of bias. Conclusions: AI models have the potential to identify edentulous areas and provide measurements of bone as part of dental implant planning using CBCT images. However, additional well-conducted research is needed to enhance the accuracy, generalizability, and applicability of AI-based approaches.