TRACKED 3D ULTRASOUND NAVIGATED LIVER BIOPSY PROCEDURE WITH REGISTERED TOMOGRAPHIC SCANS

Loading...
Thumbnail Image
Date
Authors
Schonewille, Abigael
Keyword
Biomedical Computing , Computer Assisted Surgery , 3D Ultrasound , 3D Slicer , Liver Biopsy
Abstract
Purpose: When lumps are found in the liver a liver biopsy is commonly used to help diagnose and identify the lumps. A 2D ultrasound is often used to guide such a procedure as ultrasound guidance has been shown to improve the accuracy and outcome. Unfortunately, using a 2D ultrasound to guide a biopsy needle is difficult as it can be challenging to keep the needle in the single plane of the 2D ultrasound. By using a 3D ultrasound instead this issue can be eliminated as a volume is imaged instead of a singular plane. To integrate a tracked 3D ultrasound into a liver biopsy procedure, certain criterion must be met: navigation accuracy must be sufficient for surgical use, intra-operative time must not be greatly increased, and additions to the workflow must be easy to use by the operator. The objective of this research project is to design a 3D ultrasound-guided liver biopsy workflow and software, that can be reliably used by researchers in experiments on phantoms or patients under institutional ethics. Methods: A workflow that allows for a 3D ultrasound-guided liver biopsy with the integration of pre-procedure scans was created by splitting the workflow into two treatments. A custom software was developed to allow for the easy completion of all tasks in the workflow in a streamlined form. The workflow and software were tested and validated on established test data and recorded phantom data. All calibration and registration results were then compared to ensure accuracy. The software was evaluated to ensure that it included all necessary functions for the designed workflow. Results: The average FRE error for calibrating the 3D ultrasound probe was 1.1(± 0.1) mm with an average error between the calibrations of 1.0 (± 0.4) mm. The average FRE error of the three CT registrations was 0.9 mm with a target registration error of 13.2 mm for the one verified test set. The developed software system has shown to meet the needs for the intended protocol based on the proof-of-concept testing conducted. Conclusion: In conclusion, this study has created a prototype software which effectively enables the use of a 3D ultrasound to guide a liver biopsy procedure. This was shown from the results of the various tests conducted on the workflow and software. The software shows promise for use in this application by researchers under institutional ethics, however it could benefit from additional testing and workflow changes to ensure that the TRE of the CT registration remain robustly within acceptable ranges.
External DOI