Application Effect Study of a New Training Model Combining Extended Reality Technology and 3D Printing Models in Percutaneous Balloon Compression-Accurate Puncture of the Foramen Ovale

Objective Owing to the complex structure of the skull base, accurate puncture of the foramen ovale in percutaneous balloon compression (PBC) is critical for surgical success. In traditional training models, trainees often struggle to meet the needs of precise training because of the lack of real operational simulation scenarios. This study aims to explore the application feasibility and effectiveness of a new training model that combines extended reality (XR) technology and 3D printing models in precise training for PBC foramen ovale puncture. Design Thirty participating physicians were randomly divided into a traditional group (n = 15) and a study group (n = 15). The traditional group received CT image interpretation by attending physicians with PBC experience, anatomical structure explanation of the foramen ovale using skull specimens, teaching of puncture point localization, path planning, and operation key points combined with Hartel's anterior localization method, followed by puncture training on a 1:1 ratio 3D printed model. The study group, on the basis of theoretical explanations, used XR technology for preoperative anatomical structure visualization simulation and puncture path planning, followed by hands-on practice with 3D printed models. Indicators such as the first successful puncture time and number of punctures were recorded, and training effects were comprehensively evaluated through operation scores and questionnaires. Participants The study included 30 resident physicians working and training in the Department of Neurosurgery at Jiangmen Central Hospital. Results The first successful puncture time (1.90 ± 0.15 min) and number of punctures (1.33 ± 0.13) in the study group were significantly greater than those in the traditional group (6.33 ± 0.61 min, 2.27 ± 0.25 times, P < 0.001). All physicians in the study group successfully completed foramen ovale localization and puncture operations and accurately identified the anatomical structures of the model ( P  < 0.05). The localization accuracy score (5 points, IQR 5–5) and comprehensive operation score (5 points, IQR 5–5) of the study group were significantly higher than those of the traditional group (3 points, IQR 2–4; 4 points, IQR 1–4, P  < 0.001). There was no significant difference in clinical operation confidence between the two groups before training ( P  > 0.05). After training, the clinical confidence of the physicians in the study group significantly improved ( P  < 0.05). One hundred percent of the physicians in the study group recognized that this pattern promoted learning interest and self-confidence. Conclusion A training pattern that combines XR technology and 3D printing models can strengthen trainees' spatial understanding of anatomical structures through virtual-real fusion interactive operations, shorten the learning curve, increase trainee precision and psychological confidence in clinical practice, and provide an innovative solution for standardized training involving minimally invasive interventional techniques.