Sensitivity evaluation of ArcCheck in detecting leaf open time errors of helical tomotherapy delivery / 中华放射肿瘤学杂志

ABSTRACT

Objective:To evaluate the sensitivity of the ArcCheck dosimetry system in detecting the leaf open time errors during the center and off-center helical tomotheray delivery quality assurance (DQA).Methods:Nine nasopharyngeal carcinoma (NPC) patients were selected in this study. Two DQA plans were created for each patient the" center" plan was created by moving the image of the ArcCheck phantom to place the high dose region on the phantom center and the " off-center" plan was created by offsetting the phantom and putting several diodes through a higher does region. Leaf open time errors of 2, 4, 6, 8 and 10 ms were introduced to the Sinogram which was modified using Matlab. Each intentional error plan and original (no error) plan for each patient were measured using both " center" and " off-center" DQA methods, the γ analysis was performed to evaluate the DQA results. The different dose and distance error criteria of 3%/3 mm, 3%/2 mm and 2%/2 mm were selected, and different thresholds of 5%, 10% and 15% were selected for γ analysis. The gradient and the minimum detectable error approach were taken to quantitatively analyze the sensitivity. The correlation between different dose distance error criteria and different thresholds was also evaluated by Pearson correlation analysis. Results:The absolute value of γ gradient of the " center" DQA plans were larger than those of the " off-center" plans in all different γ criteria (all P<0.05). The stricter the γ criteria were adopted, the more sensitive DQA results of leaf open time error were obtained. The minimum detectable error was 2 ms in all different γ criteria for the " center" DQA plan. The minimum error detectability of the " off-center" DQA plan was weaker than that of the " center" DQA plan. The γ passing rates of three different dose distance error criteria were significantly strongly correlated for the " center" DQA plan ( R2>0.9). For the " off-center " DQA plan, only the 3%/3 mm and 3%/2 mm criteria were significantly strongly correlated ( R2>0.9). Significant strong correlation was observed in the γ passing rate at different thresholds between the " off-center" and " center" DQA plans. Conclusions:The " center" DQA plan method is more sensitive than the " off-center" DQA plan method in all γ criteria, and the γ passing rates in different γ criteria are strongly correlated for the " center" DQA plan. The " center" DQA plan method is recommended.