Fully Automated Artificial Intelligence-Based Design Single Tooth-Supported Dental Crowns: A Systematic Review of...

Fully automated artificial intelligence (AI)-based dental crown design has recently emerged as an alternative to conventional workflows, including technician-assisted digital design and traditional wax-up techniques. However, the available evidence remains heterogeneous, and no clear consensus exists regarding its performance in terms of time efficiency, fit, and morphological accuracy. We aimed to systematically evaluate whether fully automated AI-based crown design provides comparable or superior outcomes compared with conventional approaches. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, we searched PubMed/MEDLINE, Embase, Web of Science, and Cochrane. All studies evaluating single dental crowns designed using fully automated AI systems and assessing outcomes such as time efficiency, marginal gap, internal fit, and/or morphological accuracy were included. Studies involving implant-supported or partial crowns, noncomparative simulations unrelated to crown fabrication, or those without a conventional comparator were excluded. The risk of bias was assessed using the QUIN (Quality Assessment Tool for In Vitro Studies) tool and Newcastle-Ottawa Scale (NOS). Evidence from the selected studies was synthesized. This study is registered in Prospective Register of Systematic Reviews (CRD420261359758). Of the 1291 records identified, 9 were included. Four studies were considered to have a low risk of bias, and four were considered to have a moderate risk of bias and the clinical study with good methodology quality. Time efficiency was consistently improved in AI-based workflows compared with conventional digital or wax-up techniques. Marginal gap and internal fit were generally comparable between AI and conventional digital workflows, with most values remaining within clinically acceptable thresholds. Regarding morphological accuracy, the results were heterogeneous, with some studies reporting superior outcomes for AI systems, whereas others favored technician-assisted digital workflows. Limitations of the available evidence include methodological heterogeneity, inconsistent definitions of outcomes such as "accuracy" and "work time," variability in software systems and evaluation protocols, and the predominance of in vitro studies conducted under idealized conditions. Overall, fully automated AI-based crown design demonstrates promising potential, particularly in reducing working time while maintaining clinically acceptable fit and morphological outcomes comparable to those of conventional digital workflows. However, the current evidence does not support systematic superiority of AI across all evaluated parameters. Further standardized and clinically oriented studies are needed to validate the routine integration of AI-driven crown design into prosthetic dental practice.