OpenEar Image Data Enables Case Variation in High Fidelity Virtual Reality Ear Surgery

Area of Science:

• Medical Simulation
• Surgical Training Technologies
• Anatomical Modeling

Background:

• Virtual reality (VR) simulation is a key tool for temporal bone surgical training.
• Current VR simulators primarily use computed tomography (CT) imaging.
• The Visible Ear Simulator (VES) utilizes high-fidelity cryosections, while OpenEar datasets combine cone-beam CT (CBCT) and micro-slicing for comparable model quality.

Purpose of the Study:

• To explore the integration of OpenEar datasets into the VES.
• To enable greater case variation in VR surgical simulations.
• To assess the implications for patient-specific modeling using CBCT data.

Main Methods:

• OpenEar datasets, comprising segmented and coregistered multimodal imaging of temporal bones, were utilized.
• Drillable bone segments and surface models of critical structures were derived.
• Realistic visualization was achieved through micro-slicing colorization, custom tinting, and texture mapping, with validation by clinical experts.

Main Results:

• Six of eight OpenEar datasets were successfully integrated into VES, including surgical procedure guides.
• Postprocessing steps, such as colorization and artifact removal, significantly enhanced model realism and quality.
• Common bone artifacts in CBCT data, absent in micro-slicing, were identified, particularly concerning dehiscences.

Conclusions:

• New anatomy models in VES version 3.5 offer enhanced case variation for surgical training, promoting faster and more transferable skill acquisition.
• The presence of bone artifacts in un-postprocessed CBCT data necessitates careful consideration for patient-specific VR simulation, surgical rehearsal, and planning.