The Influence of Artificial Intelligence–Assisted Intraoral Scanning on the Accuracy of Digital Implant Impressions: A systematic Review

Digital implant impressions obtained using intraoral scanners (IOS) have become integral to contemporary implant dentistry. Recently, artificial intelligence (AI)–assisted algorithms have been incorporated into IOS software to enhance scan alignment, compensate for stitching errors, fill mesh defects, and optimize scan-body recognition. Despite their rapid clinical adoption, the true influence of AI-assisted intraoral scanning workflows on the accuracy of digital implant impressions remains unclear and inconsistently reported. Recent in vitro investigations have demonstrated that implant impression accuracy is influenced by multiple interrelated factors, including intraoral scanning coverage, scan-body geometry, scanning strategy, scanner hardware, and the activation of AI-driven features. Studies evaluating AI-assisted complete-arch implant scanning workflows have shown that AI algorithms may improve trueness and precision in certain IOS systems, while producing neutral or even adverse effects in others. Similarly, investigations on image-guided photogrammetry and AI-driven coordinate superimposition have highlighted improvements in implant position transfer accuracy, yet with significant variability depending on workflow design and reference datasets. Additional in vitro evidence indicates that excessive intraoral scanning coverage can accumulate stitching errors, negatively affecting implant impression accuracy, whereas insufficient coverage may compromise alignment stability. Deep-learning–based models, including convolutional neural networks, have been used to predict optimal scanning coverage; however, these approaches remain scanner- and scenario-dependent. Furthermore, the design and geometry of intraoral and extraoral scan bodies have been shown to significantly influence trueness and precision, with AI-assisted scanning tools exerting device-specific effects that are not yet fully understood. Collectively, the existing literature demonstrates methodological heterogeneity in experimental design, reference standards, AI functionalities, and reported outcome measures (e.g., trueness, precision, RMS deviation, linear and angular discrepancies). While individual studies provide valuable insights, there is currently no comprehensive synthesis that systematically evaluates how AI-assisted intraoral scanning influences the accuracy of digital implant impressions across different in vitro conditions. Therefore, this systematic review aims to synthesize and critically appraise in vitro evidence on AI-assisted intraoral scanning workflows for digital implant impressions. The review will focus on implant position transfer accuracy, scan-body alignment, trueness, and precision, while examining the role of scanning coverage, scan-body design, scanner type, and AI algorithm activation. By consolidating current evidence, this review seeks to clarify the benefits and limitations of AI-assisted intraoral scanning and to provide evidence-based guidance for future research and clinical workflow optimization.