Quantitative Automated Source Position Assessment for HDR Gynecologic Brachytherapy Using Applicator-Specific 3D-Printed Phantoms
Abstract
Purpose
Accurate source positioning is essential for dose delivery in high-dose-rate (HDR) gynecologic (GYN) brachytherapy, necessitating routine applicator quality assurance (QA). This study introduces an applicator-specific, 3D-printed phantom with a quantitatively automated image-based method to measure source position displacement. The objective was to develop a reproducible annual QA methodology that enables direct, numerical assessment of source positioning while minimizing operator-dependent variability.
Methods
24 HDR GYN brachytherapy applicators (Geneva, Venezia, Utrecht, and Fletcher; Elekta) were evaluated. For each applicator set, a customized 3D-printed phantom was designed to ensure reproducible applicator placement. Each phantom incorporated applicator-specific slots and a fixed geometry for Gafchromic film placement, maintaining a known, consistent distance between applicators and the film plane. Autoradiography and CT imaging were performed to enable quantitative source position analysis. During autoradiography, applicators were inserted into their designated phantom slots, and four disposable X-SPOT skin markers (Beekley Medical) at the film corners served as fiducial markers for image registration. Radiopaque CT markers were inserted into each applicator for CT acquisition. A MATLAB-developed four-point registration algorithm was used to automatically register the CT images with the scanned film images. Source position displacement was quantitatively calculated in two dimensions as pixel-based distances relative to the expected source path and converted to physical units (mm).
Results
Quantitative analysis across all 24 applicators yielded a mean measured source displacement of 0.26 ± 0.01 mm. The mean source displacement (in mm with ± 0.01 deviation) for Geneva, Venezia, Utrecht, and Fletcher applicator sets was 0.33, 0.17, 0.35, and 0.24, respectively.
Conclusion
The proposed QA approach enables fully quantitative, automated calculation of source position displacement using applicator-specific 3D-printed phantoms and image registration. By providing direct numerical measurements rather than qualitative assessments, this method reduces operator dependence and improves reproducibility, supporting its use for routine annual QA of HDR GYN brachytherapy applicators.