(Wednesday, 8/1/2018) 4:30 PM - 6:00 PM
Room: Karl Dean Ballroom A1
Purpose: To evaluate custom beam models for a second check dose calculation system using statistically verifiable passing criteria for film analysis, DVH, and 3D gamma metrics.
Methods: Custom beam models for 9 linear accelerators for the Sun Nuclear Dose Calculator algorithm (SDC, Sun Nuclear) were tested using the AAPM-TG119 test suite (5 IMRT and 5 VMAT plans) and compared to the Analytic Anisotropic Algorithm (AAA) in Eclipse, and gafchromic film. AAPM-TG119 recommended confidence intervals were set to 95%. Film gamma criteria was 3%/1mm. Dose distributions were evaluated based on DVH metrics (e.g. D95, V20) and 3D gamma criteria (3%/1mm). These criteria were applied to clinical SRS/SBRT cases, 5 of each cranial and spine and 10 SBRT lung lesions. SDC was also compared to the AcurosXB algorithm in Eclipse to evaluate agreement in plans for patients with small, peripheral lung lesions, treated with SBRT, where greater accuracy in accounting for tissue heterogeneity may be required.
Results: SDC agreement with film for TG-119 cases, was 99.87±0.26% (3%/3mm), and 98.10±3.12% (3%/1mm). For comparisons between SDC and AAA: (a) TG119 plans computed for one example linac (beam model) showed PTV DVH deviation at 0.39±1.17%; (b) PTV 3D gamma rates were 99.99% (0.06%CL); (c) For cranial/spine lesions D95 was 0.99±0.86%, with gamma rates of 98.30% (6.50%CL); (d) For SBRT plans, D95 deviation was 6.19± 9.00% with gamma rates of 82.76% (75.65%CL). For SBRT comparisons between SDC and AcurosXB, D95 deviation was 2.93±3.90%; gamma rates were 94.41% (20.52%CL).
Conclusion: SDC shows agreement well within TG119 CLs for film and redundant dose calculation comparisons. TG119 plans established baselines for DVH and 3D gamma criteria. For small, peripheral lung lesions, SDC was in better agreement with AcurosXB than AAA, albeit for just 2 cases. Confirmation with a much larger sample size is warranted.
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email
