ESTRO 2024 - Abstract Book

S4782

Physics - Quality assurance and auditing

ESTRO 2024

Comparing off-axis Winston-Lutz test between ExacTrac® and cone-beam computed tomography in TrueBeam

Yan Kit Wah 1 , Ming Yan Lai 1,2 , Chung Hei Leung 1 , Man Cheuk Ng 1,2

1 St. Teresa's Hospital, Oncology Centre, Hong Kong, Hong Kong. 2 The University of Hong Kong, Department of Clinical Oncology, Hong Kong, Hong Kong

Purpose/Objective:

The BrainLab ExacTrac system and Varian cone-beam computed tomography (CBCT) are commonly used for imaging verification and correction in linac-based radiosurgery (SRS), while the emerging HyperArc system also utilizes CBCT for positioning verification and correction. The aim of this study was to evaluate the 3D phantom positional error and 2D field-target coincidence of different off-axis targets obtained from the ExacTrac system and CBCT using the MultiMet- WL QA ™ system (Sun Nuclear Corporation, Melbourne, Florida, USA).

Material/Methods:

MultiMet- WL QA ™ system with infra-red markers was used to perform off-axis Winston-Lutz (WL) test. The phantom contains 6 tungsten balls (isocenter and 5 off-axis targets ranging from 3 cm to 7 cm from the isocenter). MV portal Images of 15 combinations of couch, collimator and gantry angles were acquired with MLC configured 2cm x 2cm square fields. One 3D and two 2D results were compared which consisted of 1) the absolute 3D positional errors showing the absolute magnitudes of the 6D off-axis errors of the phantom relative to the radiation isocenter, 2) 2D target-based field-target coincidence showing the average 2D offsets of each target from the radiation isocentre and 3) 2D field-based field-target coincidence showing the individual field offset from the radiation isocenter for different couch/gantry/collimator combinations.

Results:

The absolute 3D positional phantom offset differences detected between CBCT and ExacTrac were statistically significant along the longitudinal, vertical and pitch directions (P < 0.05, with Bonferroni correction for multiple comparisons) and the absolute mean values found in CBCT were smaller (Fig. 1). For 2D target-based field-target coincidence, significantly smaller absolute mean shifts were also found in CBCT compared with ExacTrac (P < 0.05, with Bonferroni correction) for the isocenter and 4 off-axis targets (Fig. 2). For the 2D field-based field-target coincidence, 68/90 (76%) fields had CBCT with smaller offsets than Exactrac within which 26/68 (38%) were statistically significant (P<0.05 with Bonferroni correction). Among all the results that were statistically significant, 26/32(81%) showed that CBCT had smaller offsets than Exactrac.