Room: Exhibit Hall
Purpose: Most planning systems use some form of local optimization and require several iterations of setting dose/volume objectives along with arbitrary weights to achieve a physician intent on the particular disease presentation. The intermediate re-planning steps are, more often than not, necessary in order to achieve a high-quality treatment plan since local optimization can be susceptible to local minima. If the set of steps taken by planning personnel are synthesized then the clinical strain of IMRT and VMAT techniques could be drastically reduced. This study aims to automate the planning process by implementing a novel meta-optimization algorithm for prostate cancer treatment on top of a commercial treatment planning system (TPS).
Methods: The algorithm starts with the RTOG 0924 dose and volume constrains for the prostate and organs at risk and interacts with the treatment planning system to obtain dose and update constraints at every iteration. The algorithm iteratively updates the dose/volume constraints and weightings based both on the optimizer performance and the cost function defined for all constraints. These steps continue until all the dosimetric parameters are met. The dosimetric endpoints of the final treatment plan along with the steps needed to achieve this plan are investigated for broad and patient-specific trends.
Results: The meta-optimization algorithm successfully worked along the treatment planning system and was able to decrease the cost function to zero and meet all dosimetric constrains simultaneously with a minimal number of cycles.
Conclusion: The meta-optimization was able to synthesize the treatment planning process for prostate plans without the need of planning personnel oversight. Although both, more prostate patients and other disease sites are needed to fully test this system, meta-optimization represents an important step towards practical automated treatment planning. This algorithm can also be generalized to other treatment planning systems for maximal efficacy.
Optimization, Treatment Planning