Semester of Graduation

Summer 2022

Degree

Master of Science (MS)

Department

Physics and Astronomy

Document Type

Thesis

Abstract

Purpose: Treatment planning accuracy for spine stereotactic body radiation therapy (SBRT) varies depending on the dose calculation algorithm utilized in the treatment planning system (TPS). This project compared the end-to-end accuracy between spine SBRT plans calculated in a convolution-superposition based TPS (TPSCS) and Monte Carlo based TPS (TPSMC) with radiochromic film measurements. The hypothesis was that TPSMC would calculate the dose gradient in the critical region between the vertebral body and the spinal cord more accurately than TPSCS.

Methods: Single-fraction spine SBRT treatments following RTOG 0631 and local institutional guidelines were planned in TPSCS and TPSMC at five vertebral sites in an anthropomorphic phantom. The plans were delivered with a linear accelerator with Gafchromic EBT-XD film inserted at each site. For each plan, the TPS-calculated and film-measured anterior-posterior (AP) dose profiles through isocenter were obtained and 2D gamma pass rate (GPR) was calculated at multiple dose difference/distance-to-agreement criteria.

Results: For each TPS-film AP profile pair, the dose falloff difference and profile shift were measured posterior to the vertebral body. Each sample' normality was confirmed with a Shapiro-Wilk test, equality of variances was determined between the TPSCS and TPSMC samples with Levene's test, then an unpaired t-test was used to detect statistically significant differences between TPSCS and TPSMC (n = 5, α = 0.05). Across the five treatment sites, the mean dose falloff difference was -0.0 ± 0.8 mm (+0.7 ± 7.8%) in TPSCS and -0.7 ± 0.5 mm (-7.8 ± 6.0%) in TPSMC. The mean profile shift was -0.1 ± 0.3 mm in TPSCS and +0.0 ± 0.6 mm in TPSMC. No statistically significant differences were observed in the profile metrics between each TPS. TPSCS yielded consistently higher average GPR across the sites, with statistically significant differences found for all criteria tested except 2%/1mm and 1%/1mm. GPR in the critical region for both TPSs produced very accurate results for all criteria tested except 3%/1mm, 2%/1mm, and 1%/1mm occasionally.

Conclusion: Each TPS provided similar end-to-end accuracy in the critical region. The lower GPR in other areas of the TPSMC plans may be increased by adjusting TPS-specific settings to limit plan complexity.

Committee Chair

Fontenot, Jonas

DOI

10.31390/gradschool_theses.5586

Download

Share

COinS