Purpose: For modern radiation therapy (RT), a human planner needs to adjust planning objectives and constraints in the treatment planning system (TPS) to define the optimization problem for plan generation. Such a process is often tedious and time-consuming, while the plan quality can be affected by many factors, such as the planner’s experience and available planning time. To address this issue, we developed a RapidPlan assisted automatic scheme to improve treatment planning efficiency and quality.
Methods: The proposed scheme has two stages: Firstly, RapidPlan is employed to define the planning objectives and constraints in TPS to generate an intermediate plan. Secondly, the intermediate plan was evaluated by a set of planning criteria consisting of the safety limits and pre-defined physician’s objectives. If the intermediate plan fails to meet all the planning criteria, a set of rules summarized corresponding to the criteria are applied to adjust the objectives and constraints to improve the plan quality. This step can be repeatedly executed until a clinically acceptable plan is obtained. In this study, we considered the prostate-cancer-SBRT (4500cGy in 5 fractions) and prostate-Post-Operative-VMAT (7020cGy in 39 fractions) as two testbeds. 10 patient cases were employed for each type to evaluate the performance of the proposed scheme.
Results: For 7 of the 10 prostate SBRT, plans generated directly by RapidPlan achieved similar or better plan quality compared to clinical plans. The other three needed to proceed with the second stage, and clinically acceptable plans were successfully generated. Similar results were observed for Post-Operative VMAT planning with 6 of 10 going through the second stage.
Conclusion: The proposed Rapidplan assisted automatic TPS is effective to generate high-quality plans for prostate SBRT and Post-Operative VMAT. It can be useful for other planning tasks, and has the potential to serve as a tool for training new planners.
Not Applicable / None Entered.
TH- External Beam- Photons: Treatment planning using machine learning/Knowledge Based Planning/automation