Skip to main content
AAO-HNSF 2023 Annual Meeting & OTO Experience Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Back
Rate ePoster

Otology/Neurotology

(421) Real-time Volume Rendering to Predict Surgical Access in Cochlear Implantation

Monday, October 2, 2023
2:45 PM – 3:45 PM East Coast USA Time

    Presenting Author(s)

  • Samuel R. Barber, MD

    Fellow Physician
    Louisiana State University School of Medicine New Orleans
    New Orleans, Louisiana, United States

    • Co-Author(s)

  • RM

    Rahul Mehta, MD, MS, FRCS

    Associate Professor
    Louisiana State University Health Sciences Center Department of Otolaryngology-Head and Neck Surgery, New Orleans, LA
    Baton Rouge, Louisiana, USA

    • Senior Author(s)

  • Moises A. Arriaga, MD, MBA

    Clinical Professor of Otolaryngology and Neurosurgery
    Louisiana State University Health Sciences Center Department of Otolaryngology-Head and Neck Surgery, New Orleans, LA
    Metairie, Louisiana, United States

Disclosure(s):
Samuel R. Barber, MD: No relevant relationships to disclose.
Moises A. Arriaga, MD, MBA: No relevant relationships to disclose.


Introduction: Cochlear Implantation (CI) with Round Window (RW) insertion is performed via a transmastoid approach through the facial recess. Computed Tomography (CT) imaging aids in surgical planning, but assessing RW access is limited by standard 2D reformatted views. Commercially available surgical planning software calculates anatomic measurements such as cochlear duct length, yet there is no capability to evaluate anatomic constraints encountered during dissection and exposure through the facial recess. Volume Rendering is a technique that simulates light and can project CT imaging with any 3D perspective. Herein, we propose a novel 3D modality using a game engine that predicts RW accessibility with simulated views of surgical corridors.

Methods: A pilot study evaluated feasibility of volume rendering of CT imaging to accurately represent anatomy using virtual and physical simulation with 3D-printed patient-specific models. Subjects comprised CT temporal bone imaging with 1mm slices. Volume rendering was performed directly from DICOM data with native support in an app developed using Unreal Engine. A user interface was designed using visual scripting and C++ code. A virtual camera projected 3D anatomy from a surgical view and planar cropping of the facial recess allowed RW visualization. Physical models were manually-segmented, 3D-printed, and dissected in a training laboratory. The St. Thomas’ Hospital (STH) classification for RW accessibility was used to compare virtual and physical views.

Results: CT studies from 3 subjects rendered successfully with RW visualization through a virtual facial recess from a surgical position in Unreal Engine. 3D-printed models were drilled to expose the RW. Virtual and physical views were compared and rated to have the same STH type in all 3 subjects.

Conclusions: Volume Rendering in a game engine is feasible to accurately project imaging to predict anatomic relationships with near-instant file preparation without segmentation. Future studies will compare virtual views to intraoperative findings in CI surgery.