Date Approved


Graduate Degree Type


Degree Name

Medical Dosimetry (M.S.)

Degree Program

Health Professions

First Advisor

Jake DeWalt

Academic Year



Introduction: Inter fractional setup uncertainties for patients receiving cervical radiation therapy treatments can cause a significant change in dose to the target structures resulting in under dosing or over dosing the patient. Varian has created an auto feathering algorithm for extended field radiotherapy to help smooth out the dose located at the junction between the two fields. This may reduce the percent change to the dose distribution when the setup uncertainties are introduced.

Methods: The study is a retrospective dosimetric analysis of 10 randomly selected female cervical cancer patients. Each patient had a plan created using the traditional technique of overlapping the fields without the use of the auto feathering algorithm, and then with the auto feathering algorithm enabled. The patients were then shifted ±3 mm and ±5 mm in the CC direction and then had the dose recalculated without any changes to optimization.

Results: The results showed statistically insignificant percent changes to the planning target volume, global doses, or surrounding organs at risk. Although the percent differences were all lower on the plans that incorporated the use of the auto feathering algorithm it did not produce a large change compared to the traditional overlap technique.

Conclusion: The Varian auto feathering algorithm can help to reduce the percent changes to the dose distribution when setup uncertainties are introduced. However, the difference between the plans that used the algorithm versus the plans that did not were not significant enough to require the use of the algorithm when a quality initial overlap plan is created.