MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel

Evaluation of a Novel Automation Software for Generating Field-In-Field Plans for Various Treatment Sites

C Esquivel1*, L Patton1, B Doozan2, K Nelson1, D Boga1, T Navarro3, (1) Texas Oncology San Antonio, San Antonio, TX, (2) Texas Oncology, McAllen, TX, (3) Texas Oncology Brownsville,TX

Presentations

(Sunday, 7/12/2020)   [Eastern Time (GMT-4)]

Room: AAPM ePoster Library

Purpose: Comparison study of traditional Field-in-Field (FiF) treatment plans to an automated approach using RADformation’s software, EZFluence.

Methods: A study evaluating 50 treatment plans of varying sites including breast, whole brain and rectum were created using the traditional FiF planning. Treatment plans included mixed energy fields of 6 and 18 MV in the Eclipse TPS. EZFluence, an embedded script in Eclipse, allows the planner to automate the FiF process. The target and critical structures are based on user specification with the desired coverage reviewed prior to creation of a FiF plan in the software. Comparison to the original plan’s prescription dose coverage, maximum dose to the target and the total MU of each field were documented. Time required to create an EZFluence plan, subfield merging, and normalization was additionally recorded. Plans were validated with RadCalc and MapCheck2.

Results: EZFluence produced comparable plans in a relatively shorter timeframe. When normalized to produce the same coverage of the original plan, the dose distribution, hotspot and dose to normal tissue structures were on the average within 1% of the original plan. Total MUs increased, on average, 4.5% (13 MUs). Average hotspot to homogenous plans was 106%. RadCalc was within 5% and MapCheck2 demonstrated agreement of a passing rate of 95% (using 2%/2cm/10). Average time commitment for the creation of FiF plans through traditional steps was 10-20 minutes. With EZFluence, time was greatly reduced to 4-9 minutes.

Conclusion: EZFluence achieves comparable plans (within 1%) to traditional FiF planning and demonstrates a significant reduction in the time with the Field-in-Field planning process. Dosimetric evaluation with RadCalc and MapCheck confirm accuracy and feasibility of EZFluence within the clinical environment.

Keywords

Treatment Planning

Taxonomy

TH- External Beam- Photons: Treatment planning using machine learning/Knowledge Based Planning/automation

Contact Email