Comparison of brain planar kV and CBCT imaging modalities for positional accuracy using 6DoF couch
Stefano Sirianni,
United Kingdom
PD-0652
Abstract
Comparison of brain planar kV and CBCT imaging modalities for positional accuracy using 6DoF couch
Authors: Stefano Sirianni1, Christopher Dean2, Alison Cole2, Liam Wiles1, Rachel Lewis1
1Barts Health NHS trust, Radiotherapy, London, United Kingdom; 2Barts Health NHS trust, Physics, London, United Kingdom
Show Affiliations
Hide Affiliations
Purpose or Objective
- To compare 2D planar kV image registration results against three-dimensional cone beam CT (CBCT) for patients undergoing radical radiotherapy to the brain, using 6 Degrees of Freedom (6DoF) couch corrections on Varian™TrueBeam.
- To quantify the extent of residual errors following the application of 6DoF couch corrections.
Material and Methods
5 patients undergoing radiotherapy to the brain were immobilised using Civco® Uni-frame IMRT thermoplastic shells with standard clear plastic headrest.
Patients were first imaged with a kV orthogonal pair; these images were auto-registered in the Varian™ 2D/3D matching workspace where shifts were recorded but not applied. A CBCT scan was then acquired, auto-registered and 6DoF shifts applied prior to treatment.
A second CBCT was taken after the applied couch shifts but prior to treatment to detect whether the patient had counteracted the applied corrections.
Results
142 CBCT images and 142 kV planar image pairs were obtained and analysed. The mean ± S.D. differences between kV planar and CBCT registration results were: VERT 0.0 ± 0.3 mm, LONG 0.0 ± 0.4 mm, LAT -0.2 ± 0.4 mm, RTN 0.0° ± 0.2, PITCH -0.1° ± 0.2, ROLL 0.0° ± 0.2. The mean absolute differences between the registration results of the two imaging modalities were all less than 0.4 mm in translation and less than 0.2 degrees in rotation.
The mean ± S.D. CBCT registration results following application of couch corrections were: VERT -0.2 ± 0.2 mm, LONG 0.0 ± 0.3 mm, LAT -0.2 ± 0.2 mm, RTN 0.1° ± 0.2, PITCH 0.1° ± 0.2, ROLL 0.0° ± 0.1. The mean absolute differences were all less than 0.3 mm in translation and less than 0.2 degrees in rotation.
Conclusion
The high degree of concordance between the 2D planar and CBCT registration results confirm that the modalities can be used interchangeably. The small residual deviations following couch corrections indicate that in general, patients did not counteract their application. However, for a small number of fractions larger deviations were detected after application of the corrections. In these cases, shell fit was found to be sub-optimal due to patient contour change. This was prevalently seen towards the end of the treatment course and corrected by improving shell fit and reducing patient movement within the shell.
This study has shown that it is possible to significantly reduce the use of daily CBCT for this cohort of patients whilst maintaining excellent levels of patient positional accuracy. The replacement of CBCT with 2D planar kV in this institution has resulted in reduced concomitant doses and shorter overall treatment . This study has also highlighted the importance of good immobilisation and the continuous assessment of shell fit throughout radiotherapy. We recommend the use of weekly CBCT imaging for 3D contour evaluation for dosimetric assessment.