Background and Aims:
Surgical excision of abdominal neuroblastomas (NB) presents considerable anatomical complexities, particularly when multiple image-defined risk factors (IDRFs) are present.
This research introduces a prototype 3D NB model integrated into a Virtual Reality (VR) application, tailored for junior doctor education and surgical planning.
Methods:
• A 3D Model was digitally reconstructed from a patient’s dataset (including computed tomography (CT) and magnetic resonance (MR) images) and visualized in VR through an Oculus Quest 2.
• Interactions in VR allow for CT dataset overlay, toggling individual structures visibility, 360-degree navigation, providing information and the ability to annotate.
• User testing at the Royal Hospital for Children, Glasgow, involved 20 medical professionals of varying levels of training, exploring three aspects through mixed methods research : (1) A counterbalanced design contrasting two-dimensional (2D) versus VR identification of anatomical structures (Mann-Whitney U test and Wilcoxon-signed rank analysis), (2) Usability testing via the System Usability Scale (SUS), and (3) Presence assessment via the ITC-Sense of Presence Inventory (ITC-SOPI) questionnaire.
Results:
Fewer errors in anatomical identification were present in VR compared to 2D (p=<0.05), with a higher mean error value if 2D was used prior to 3D (1.2 vs 0.5).
SUS ranked highly (79.75), and all presence constructs were positively scored: Spatial (Mean 3.75 ± 0.55 SD), Engagement (Mean 4.08 ± 0.4 SD), Ecological Validity (Mean 3.72 ± 0.83 SD), and Negative Effects (Mean 1.77 ± 0.78 SD).
Conclusion:

This 3D neuroblastoma model in VR demonstrated improved anatomical recognition, as a user-friendly tool to enhance both surgical planning and medical education. This provides a foundation for future research and development within the context of comparable rare tumour masses.