Room: Room 202
Purpose: We recently developed high grid ratio 2D antiscatter grid prototypes, and demonstrated that HU uniformity and accuracy in CBCT is significantly improved using 2D grids. In this study, we investigate the effect of 2D grid on CBCT based dose-calculation accuracy for monitoring dose delivery during treatment delivery.
Methods: CBCT images were acquired on a linac mounted CBCT system using a high grid ratio 1D antiscatter grid and with our 2D ASG prototype with a grid ratio of 12. These images included a CT electron density (CTED) phantom and an anthropomorphic pelvis phantom (â€œstandardâ€? and â€œlargeâ€? configurations). For dosimetric comparisons, gold-standard phantom images were acquired with a multi-detector CT. A CTED table was generated for each grid configuration, and Acuros dose calculation algorithm was employed. Clinically relevant pelvic SBRT treatment plans were generated in both soft-tissue and bony anatomy regions using gold standard images, and then plans were recalculated on the CBCT images acquired with both grids.
Results: For all treatment sites and dosimetry metrics, the 2D grid prototype provided better accuracy in dose calculations than the 1D grid. The largest dosimetric deviations for the 1D grid were for an L5 vertebral body SBRT treatment plan, observed in the â€œlargeâ€? pelvis phantom. In both the target region and proximal normal tissues, the dose was overestimated by 5%, compared to the gold standard, whereas dosimetric deviations with the 2D grid was less than 0.57% in the large pelvis phantom. The lowest deviation for the 1D grid was 1.7% for a centrally located soft-tissue target.
Conclusion: When compared to existing 1D grid technology, 2D grids can significantly improve dose calculation accuracy in CBCT images. Our results indicate that the use of 2D grids in linac mounted CBCT systems may be a viable solution for monitoring dose delivery accuracy during radiotherapy treatments.
Funding Support, Disclosures, and Conflict of Interest: Acknowledgement: This project is supported in part by NIH/NCI R21CA198462. Tesla K40 GPU was provided by NVIDIA Corporation. CBCT electron density phantom was provided by Gammex Inc.
Not Applicable / None Entered.