-
Notifications
You must be signed in to change notification settings - Fork 102
Plan Robustness Analysis
DVH robustness analysis tool simulates the robustness of dose distribution to setup uncertainties. It requires estimates of patient setup errors between fractions. These setup uncertainties comprise of systematic and random components. Here are couple of references for estimating patient setup uncertainties.
Kupelian et al, “Multi-institutional clinical experience with the Calypso System in localization and continuous, real-time monitoring of the prostate gland during external radiotherapy”, International Journal of Radiation OncologyBiologyPhysics, Volume 67, Issue 4, 2007, Pages 1088-1098, ISSN 0360-3016, https://doi.org/10.1016/j.ijrobp.2006.10.026.
Munck af Rosenschöld et al, “Modeling positioning uncertainties of prostate cancer external beam radiation therapy using pre-treatment data”, Radiotherapy and Oncology, Volume 110, Issue 2, 2014, Pages 251-255, ISSN 0167-8140, https://doi.org/10.1016/j.radonc.2013.12.010.
The tool supports specification of setup errors as translations in x (L-R), y (A-P) and z (S-I)-directions. The tool assume that the shifts are normally distributed. Hence, it requires mean and standard deviation (sigma) for inter and intra-fraction motion.
Inter fraction mean = mean of all fractions and patients per direction (LR, AP or SI) Inter fraction sigma = standard error of the above mean per direction (LR, AP or SI) Intra fraction mean = generated by sampling from inter fraction distribution above. Intra fraction sigma = standard deviation of fractions averaged across patients.
For each trial, the patient (CT and segmentation) is shifted by the following amount: Shift = fraction mean + inter fraction sigma (normally drawn) + intra fraction sigma (normally drawn)
Voxel-wise dose is accumulated for each fraction which sums up to the total dose for that trial. Number of fractions is specified as an input parameter. This process is repeated for the specified number of trials. Number of trials is specified an an input parameter.
In order to visualize the 1/2/3-sigma DVH ranges, the following equation is used to compute dose at each voxel: (Mean of all trials) +/- 2*(sigma for all trials)
DVH is then generated for the upper and lower bounds. GUI displays statistics such as min/max/mean and Vx.