Utilization of a novel radio-opaque 3D printed phantom for simulation training of CT-guided interventional procedure

Background: Training radiology residents to perform CT-guided procedures can be challenging due to limited access to scanner time, competition for procedures with other residents and fellows and the risk of exposing patients to unnecessary additional radiation due to the inexperience of the primary operator. Simulation training has shown great benefit in reducing patient risk and increasing training opportunities. The aim of this study was to assess the impact of a high-fidelity 3D-printed model as an interventional radiology training tool for trainees. Methods: This study was approved by the local research ethics committee. Using anonymised patient scan data from a prone CT colonography study, the pelvis and two vertebrae were 3D printed in an experimental radiopaque resin. A five-sided mould was created for casting the pelvic phantom. The lower back soft tissue impression was converted to a surface using Geomagic software. This surface was subtracted from a solid base using a Boolean extraction. Gel wax was gently poured into the mould to reduce air bubble formation. The phantom was allowed to cool and solidify overnight. Radiology trainees at our institution were invited to participate in the study and received formal teaching by a musculoskeletal fellowship-trained radiologist before performing two independent CT-guided needle placements on the model. Data were recorded on trainee confidence prior to and after the training session, time to complete the procedure and dose to the model for each attempt and compared using paired t-tests. Results: Twelve radiology trainees took part in the study. Mean total procedure time and dose to the model decreased significantly for the second attempt compared to their first attempt at needle placement in the SI joint (6.9 vs. 4.2 min, p = 0.008, DLP 31.1 vs. 25.9, p = 0.022 respectively). Trainees were significantly more confident in performing a CT-guided SI joint injection following the training session. Conclusion: This study demonstrates that a 3D-printed model hip can provide a valuable training experience for radiology trainees for performing CT-guided interventions and can improve performance and confidence, without conferring any risk to patients. This model could be incorporated into radiology training curricula.