Encrypted login | home

Program Information

An Assessment of Deviations in Gamma Index Between PTW Octavius4D Phantom Measurements and Various Treatment Planning Systems

J Roring

JE Roring*, A Alexandrian , N Papanikolaou , S Stathakis , University of Texas HSC SA, San Antonio, TX


SU-F-T-280 (Sunday, July 31, 2016) 3:00 PM - 6:00 PM Room: Exhibit Hall

Purpose: There is a variety of treatment planning systems (TPS) available for clinical use all with highly variable settings to calculate patient dose. Patient specific IMRT QA is often a daily procedure at most clinics and each clinic devises their own policy on the TPS settings for testing the quality of the patient plan against a measured dose on IMRT QA phantoms. With this in mind, it is the aim of this study to evaluate the effect of dose calculation parameters on IMRT QA.

Methods: Six patient plans were optimized and calculated using Pinnacle³ (v9.10). Patient QAs were calculated using the PTW Octavius4D phantom with a dose grid resolution of 3mm and a phantom density of 1.045g/cc. Using PTW’s VeriSoft, the calculated dose distribution of Pinnacle³, Monaco, and Eclipse were compared to the delivered plan (the control) on the Octavius4D. For each TPS, the dose grid was varied from 2-4mm, and the phantom density varied ±10% in increments of 2%. Additionally, statistical uncertainties ranging from 0.7%-3% were calculated for Monaco. Using the evaluation metric of 3% and 3mm with a threshold of 10% of the max dose, the average and standard deviation of the 3D γ-index were calculated for each plan. The γ-index from axial, sagittal, and coronal slices at isocenter and the overall 3D γ-index were used for the comparisons.

Results: The disparity in γ-index was small between most plan modifications. An extreme example is one case in which a 3D slice yielded a γ-index that differed up to 12.9% whereas the maximum variation in total volume for any case was merely 2.3%. The average standard deviation of all slices was ±1.49% and the average standard deviation for total volume was only ±0.55%.

Conclusion: Dose calculation resolution and algorithm does not significantly affect the patient QA measurements.

Contact Email: