Program Information
A Clinically Relevant IMRT QA Workflow: Design and Validation
C Stambaugh*, G Ezzell , Mayo Clinic Arizona, Phoenix, AZ
Presentations
TH-AB-FS1-1 (Thursday, August 3, 2017) 7:30 AM - 9:30 AM Room: Four Seasons 1
Purpose: To determine clinically relevant pass/question/fail criteria for gamma analysis of IMRT QA plans, identify which plans should be further analyzed with DVH metrics, and create a workflow for performing that DVH-based analysis.
Methods: Eleven plans, prostate and head/neck, were selected based on their high passing rate under the current gamma criteria. These were modified by moving the MLCs to underdose the target or overdose important structures. Commercially available hardware/software was used to measure and analyze all plans (76 total) using 4%/3mm, 3%/3mm, 3%/2mm and 2%/2mm criteria. Two ROC curves per criterion were created to assess effective passing rates. One ROC curve was for a higher threshold that determined a pass and the second denoted a failure. Plans between these two thresholds need DVH-based analysis. A 3D dose reconstruction was also performed on all plans to access the accuracy and usefulness of DVH-based analysis.
Results: Analysis of the ROC curves showed optimal pass and fail thresholds for plan error detection per criterion. Based on measurement uncertainty and pass/fail ranges, 3%/2mm gamma criteria with a pass level of 95% and a fail level of 90% was most optimal. DVH analysis showed good agreement with the all reconstructed plans except where D95% of the target was significantly cold. For questionable plans, comparing the organ-specific DVHs to the planning constraints proved to be an efficient and effective workflow.
Conclusion: This work indicates the potential for significant improvement in error detection for IMRT QA. Using effective pass/fail thresholds to determine plans that need DVH-based analysis minimizes the need for excessive analysis and making use of the dosimetric constraints of the plan minimizes the burden on physicians. Overall, DVH-based analysis is a powerful tool that can provide significant insight over the traditional approach.
Contact Email: