Room: Stars at Night Ballroom 2-3
Purpose: To develop a versatile 3D ultrasound (US) system for needle guidance during high-dose-rate interstitial brachytherapy (ISBT) of gynecologic cancers, providing the potential for intraoperative implant assessment and improved implant quality, reducing risk to nearby organs-at-risk (OAR).
Methods: To address the diverse presentations of gynecologic malignancies, three 3D US modes were incorporated into the system: transrectal US (TRUS), 360Â° sidefire transvaginal US (TVUS), and endfire TVUS. All 3D images were acquired in < 20 s by rotating a 2D US probe. Patient studies have been completed for the TRUS and 360Â° TVUS modes, including 58 and 54 needles, respectively. Endfire TVUS was tested in a phantom with 12 needles. Needle positions were assessed relative to post-insertion CT images, reflecting the clinical standard. To improve the systemâ€™s clinical utility, automatic needle segmentation leveraging the randomized 3D Hough transform was implemented to identify multiple needles simultaneously and tested on a 360Â° TVUS image, as a proof-of-concept.
Results: Nearby OAR were clearly visualized in patient images, particularly the bladder in TRUS and 360Â° TVUS and the rectum in 360Â° TVUS images. The mean Â± standard deviation needle positional differences were 3.82 Â± 1.62 mm, 2.36 Â± 0.97 mm, and 1.91 Â± 0.24 mm for TRUS, 360Â° TVUS, and endfire TVUS (phantom), respectively. The mean angular differences were 3.04 Â± 1.63Â°, 1.95 Â± 0.70Â°, and 1.51 Â± 0.81Â°. All eight inserted needles were identified using the segmentation algorithm with a mean positional difference of 0.78 Â± 0.17 mm and angular difference of 0.44 Â± 0.19Â°, compared to manual segmentations.
Conclusion: The 3D US system has the potential to be adopted intraoperatively during gynecologic ISBT procedures to assess the interstitial implant with both needles and nearby OAR visible, providing versatile scanning options to account for the variability to gynecologic cancer presentations.
Funding Support, Disclosures, and Conflict of Interest: This work was funded by the Ontario Institute of Cancer Research (OICR), the Canadian Institutes of Health Research (CIHR), and the Natural Sciences and Engineering Research Council of Canada (NSERC). We have no other conflicts of interest or disclosures.