Program Information

Experience-Based VMAT Plan Quality Database

K Bernstein¹*, N Kalach² , B Wolthuis³ , C Tai⁴ , A Kravchuk⁵ , E Bernstein⁶ , (1) Mount Sinai Beth Israel, New York, NY, (2) Mount Sinai Beth Israel, New York, NY, (3) Mount Sinai Beth Israel, New York, NY, (4) Mount Sinai Beth Israel, New York, New York, (5) Brooklyn Technical High School, Brooklyn, New York, (6) Stuyvesant High School, New York, New York

Presentations

SU-E-T-413 Sunday 3:00PM - 6:00PM Room: Exhibit Hall

Purpose: To quantify VMAT plan quality using a retrospective study of over 200 clinical treated VMAT plans created using the Eclipse Treatment Planning System to create benchmarks of plan quality for a few categories of treatment sites.

Methods: Using a controlled phantom geometry, various dosimetric indices were investigated to quantify dosimetric plan quality as a function of isocenter displacement from center of mass, average path length, number of arcs and PTV proximity to critical structures. Beginning with published dosimetry indices from SRS and SBRT evaluations, UDI (Unified Dosimetry Index) and modified UDI were tested before creating a new factor VMAT-DI. VMAT-DI was developed within boundaries of this project and it includes renormalized factors of conformity index, coverage index, modified gradient index and homogeneity index as well as indices based on routine clinical practice such as absolute dose max index. The plans were then evaluated using the VMAT-DI such that benchmarks for planning could be created.

Results: The majority of the plans evaluated could be assigned VMAT-DI values within a range for each treatment site. However, the outliers were results of difficult planning parameters such as very irregular targets, inhomogeneities or difficult to achieve critical structure constraints. To effectively use VMAT-DI for guidance, especially for prediction of the plan quality for body sites new to the practice, VMAT-DI database needs to be subdivided by target complexity and by body site index/average path length factor.

Conclusion: An experienced-based VMAT-DI database can be used to help analyze plans before evaluation by the physician to show that it adheres to the clinical standards of previously treated VMAT plans which will make a guideline for concluding the optimization. The introduction of institution-wide clinical planning protocols, standardizing OAR naming and constraints will make it possible to incorporate a cumulative critical structure dosimetry index such as NTCP.

Contact Email: