PhD Scientific Days 2024

Surgical Medicine

Semi-Automatic Method for the Design of 3D Printed Pedicle Screw Guides

Text of the abstract

Introduction: Accurate placement of pedicle screws is imperative in spinal surgery to ensure biomechanical stability and successful fusion. Intraoperative planning and execution of this procedure can be prone to errors and involve excessive ionizing radiation. The use of 3D printed pedicle screw guides is an established and cost effective method for intraoperative navigation but their design can be time consuming. We present a semi-automatic method for designing 3D printed pedicle screw guides, aiming to enhance precision and efficiency in spinal surgery.
Aims: Our objective is to develop a semi-automatic method for the design of 3D printed pedicle screw guides. We aim to assess the feasibility, accuracy, and efficiency of this approach.
Methods: We employ preoperative CT scans of patients undergoing primary spinal surgery to extract pedicle screw axes' endpoints and vertebral anatomy. Using AI-based segmentation algorithms, we automatically delineate vertebrae. A Python script was developed to generate drill guide templates based on 3D point coordinates and vertebral geometry. The generated template geometry and vertebral geometry are then 3D printed for physical implementation. Finally, a test surgery is performed to evaluate the usability and precision of the guide template.
Results: Our method demonstrates feasibility and efficiency, with a processing time of less than 10 minutes per vertebra. AI-based vertebra segmentation achieves accurate delineation, reducing manual labor and potential errors. The algorithmically generated drill guide templates fit precisely to the corresponding vertebral anatomy. Physical implementation via 3D printing results in tangible guides that enables the user to insert pedicle screws in a way that aligns closely with preoperative plans.
Conclusion: The presented semi-automatic method offers a significant advancement in designing pedicle screw guides for spinal surgery. By integrating advanced imaging, AI, and algorithmic design, we achieve a streamlined approach that prioritizes precision and efficiency. Rapid processing time and accurate customization enhance surgical workflow and patient outcomes, representing a promising avenue for future research and clinical implementation.