ARTIFICIAL INTELLIGENCE AS A DECISION-SUPPORT TOOL FOR MATERIAL SELECTION IN PROSTHODONTICS: AN ANALYTICAL REVIEW OF THE LITERATURE

Abstract

Background. The rapid development of digital dentistry and artificial intelligence (AI) is expanding the range of technologies available for diagnosis, treatment planning, prosthetic design, and outcome prediction. Material selection remains a clinically important component of prosthodontic treatment because the choice of restorative material must account for mechanical, functional, biological, and esthetic requirements. Aim. To analyze recent literature on the application of AI in prosthodontics, with particular attention to its potential role in supporting restorative material selection, integrating clinical and digital parameters, predicting prosthetic outcomes, and personalizing treatment planning. Materials and methods. A narrative analytical review was developed from literature published during seven completed calendar years (January 2019–December 2025). Relevant publications were identified in PubMed/MEDLINE, Scopus, Web of Science, and NCBI resources using combinations of terms related to artificial intelligence, machine learning, prosthodontics, dental materials, CAD/CAM, crowns, dental prostheses, occlusion, and treatment outcomes. Publications were thematically analyzed according to AI application, type of prosthodontic task, data source, clinical purpose, reported outcomes, and limitations. Results. The reviewed literature demonstrates that AI is increasingly applied in prosthodontic diagnostics, treatment planning, automated crown design, finish-line detection, occlusal assessment, prediction of prosthetic outcomes, and digital workflows. AI-assisted CAD systems have shown potential to improve design efficiency and, in selected studies, achieve clinically acceptable crown fit. However, the evidence remains heterogeneous with substantial differences in datasets, algorithms, validation procedures, and outcome measures. Direct evidence supporting independently validated AI systems for individualized restorative material selection remains limited. Current evidence therefore supports the use of AI primarily as a clinical decision-support tool rather than as an autonomous system for material selection. Conclusions. AI has considerable potential to support individualized material selection and prosthodontic treatment planning by integrating multiple clinical and digital parameters. At the same time, current evidence does not support treating AI as an autonomous replacement for professional clinical judgment. Direct validation of AI systems specifically designed to select restorative materials remains limited, and many predictive models require stronger external validation and prospective clinical evaluation. The most appropriate current framework is therefore a human-in-the-loop decision-support model in which AI augments, rather than replaces, clinical expertise