Advancements of artificial intelligence algorithms in predicting dental implant prognosis from radiographic images: A systematic review.

STATEMENT OF PROBLEM

The ability of artificial intelligence (AI) to accurately forecast the prognosis of dental implants from radiographic images is unclear.

PURPOSE

The purpose of this systematic review was to evaluate the efficacy of AI algorithms in predicting implant outcomes by focusing on key factors like peri-implantitis, implant stability, marginal bone levels, dental implant failure, implant success, and osseointegration.

MATERIAL AND METHODS

This systematic review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses of Diagnostic Test Accuracy (PRISMA-DTA) guidelines. The included studies focused on the radiographic data of patients with dental implants where AI algorithms were compared with expert judgment. A comprehensive search in 4 databases and a manual search were conducted. The quality and risk of bias were assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool.

RESULTS

Of 424 references, 13 eligible articles were included. These studies used different radiographic types and AI models. AI algorithms showed promising accuracy rates, reaching 99.8%. Sensitivity and specificity ranged from 67% to 95% and 78% to 100%, respectively. The studies indicated that AI models significantly reduce analysis time compared with manual methods.

CONCLUSIONS

AI algorithms demonstrate promising accuracy in predicting dental implant prognosis, enhancing treatment planning, and early intervention. However, variations in AI models and methodologies highlight the need for further research.