Early refractive stability and predictive accuracy of a novel hydrophobic C-loop monofocal intraocular lens in eyes with normal axial length.

The integration of virtual reality (VR) and artificial intelligence (AI) into medical education represents a transformative shift in how learner performance is assessed and supported. VR-based simulations provide standardized, immersive environments for evaluating both technical and non-technical skills, while AI-driven systems now enable automated scoring, personalized feedback, and competency tracking. Recent studies demonstrate the feasibility of AI-powered formative feedback systems in undergraduate medical education, with high levels of learner and mentor acceptance, as well as measurable improvements in structured, rubric-based feedback supported by generative AI. In simulation contexts, AI-assisted assessment of non-technical skills shows partial agreement with expert raters, indicating potential for hybrid human–AI evaluation models. Immersive technologies are increasingly incorporated into patient safety and anesthesiology training to enhance competency-based assessment frameworks. At the same time, concerns remain regarding the validity, reliability, and safety of AI-supported assessment. Evaluations of large language models on medical examination-style questions reveal strong performance in text-based tasks but reduced reliability in image-based interpretation, raising important questions about assessment robustness. Similarly, while AI-based chatbots providing real-time communication feedback demonstrate high perceived accuracy and usefulness, further validation is required before broad curricular integration. These findings underscore both the promise and the methodological challenges associated with integrating VR and AI into high-stakes educational evaluation. Although publications on VR- and AI-enabled medical education have increased markedly in recent years, the intellectual structure and developmental trajectory of research specifically focused on educational evaluation remain unclear. Existing studies are often technology- or specialty-specific and lack a comprehensive synthesis of thematic evolution and research frontiers. Therefore, this study conducted a bibliometric analysis to map publication patterns, collaboration networks, and knowledge structures in VR- and AI-based medical education evaluation. By delineating the temporal and thematic dynamics of this emerging field, the study aims to provide an evidence-based foundation for future research and the development of more scientific and standardized evaluation frameworks.