- 251 views
- 423 downloads
3D Ultrasound Guidance System for Pedicle Screw Insertion in Scoliosis Surgery
-
- Author / Creator
- Chan, Andrew Y-H
-
This thesis describes the research and development process in creating a 3D ultrasound-based navigation system to assist screw insertion for scoliosis surgery, with a focus on children who have adolescent idiopathic scoliosis (AIS). Screw placement accuracy is critical to prevent spinal cord or other neurologic injury in spine surgery. Traditional image guidance can reduce pedicle breach rates in posterior fusion surgery for AIS, but at the expense of added ionizing radiation to patients and staff, prolonged surgeries and hindrances to surgical workflow.
A new image guidance system was developed by combining motion capture cameras, conventional medical ultrasound, and image registration to display navigation information in a three-dimensional virtual environment. In this system, motion capture markers are mounted onto the medical ultrasound transducer to pair 2D images with motion capture information, allowing for a 3D reconstruction of the vertebral surface. A pre-operative 3D model of the vertebra is then image registered to the 3D ultrasound vertebral surface to localize the vertebra in the operating room. The vertebra is then display on a screen, alongside motion captured surgical tools to allow surgeons to have visual feedback on the entry-point and trajectory of their screw placements.
The system was evaluated for accuracy and speed in four stages: motion capture was evaluated first, followed by 3D ultrasound localization accuracy. The image registration speed and accuracy were then evaluated, and lastly the screw placement accuracy was evaluated. This research focused on using 3D-printed ultrasound phantoms to evaluate accuracies, alongside a pilot study on porcine spines.
The accuracy of the motion capture cameras was evaluated, with translational accuracies of 0.25mm for translations within 15mm and rotational accuracies within 3.7o for rotations within 60o. The accuracy of 3D ultrasound in localizing landmarks was found to be 0.8±0.6mm, while using a motion capture probe to localize landmarks on the phantom found a positional error of 1.1±1.1mm and rotational error of 0.0±1.7o. The accuracy of the image registration algorithm was 0.3±0.2mm and 0.9±0.8o, while the surgical localization accuracy was 1.2±0.5mm and 2.2±2.0o. A qualitative study on a porcine spine showed that image registration was successful for 91% of the porcine registrations.
Lastly, the navigation accuracy of the system was tested using a live motion capture probe, which found a final entry-point and trajectory errors of 0.5±0.3mm and 2.0±0.8o, with 95.6% of screw placements meeting an entry-point target accuracy of 1mm, and all screw placements meeting a trajectory target of 5o. The average processing time was 8.9±1.4s.
This system has the speed and accuracy for usage in spine surgery. However, further evaluation needs to be completed before clinical evaluation. First, improvements to the user interface and integration into surgical workflow needs to be tested. Second, the most appropriate pre-operative imaging for the system needs to be selected, whether it is X-rays, computed tomography or magnetic resonance imaging. Lastly, quantitative evaluation of the ultrasound system on soft tissues including porcine spines or cadaver spines, needs to be completed.
-
- Subjects / Keywords
-
- Graduation date
- Fall 2020
-
- Type of Item
- Thesis
-
- Degree
- Doctor of Philosophy
-
- License
- Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.